Cart (Loading....) | Create Account
Close category search window
 

Robotics:Science and Systems IV

Cover Image Copyright Year: 2009
Author(s): Brock, O.; Trinkle, J.; Ramos, F.
Publisher: MIT Press
Content Type : Books & eBooks
Topics: Robotics & Control Systems
  • Print

Abstract

Robotics: Science and Systems IV spans a wide spectrum of robotics, bringing together researchers working on the foundations of robotics, robotics applications, and analysis of robotics systems. This volume presents the proceedings of the fourth annual Robotics: Science and Systems conference, held in 2008 at the Swiss Federal Institute of Technology in Zurich. The papers presented cover a range of topics, including computer vision, mapping, terrain identification, distributed systems, localization, manipulation, collision avoidance, multibody dynamics, obstacle detection, microrobotic systems, pursuit-evasion, grasping and manipulation, tracking, spatial kinematics, machine learning, and sensor networks as well as such applications as autonomous driving and design of manipulators for use in functional-MRI. The conference and its proceedings reflect not only the tremendous growth of robotics as a discipline but also the desire in the robotics community for a flagship event at which the best of the research in the field can be presented.

  •   Click to expandTable of Contents

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Front Matter

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): i - xv
      Copyright Year: 2009

      MIT Press eBook Chapters

      This chapter contains sections titled: Half Title, Title, Copyright, Contents, Preface, Organizing Committee, Program Committee, Sponsors View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Multi-Sensor Lane Finding in Urban Road Networks

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 1 - 8
      Copyright Year: 2009

      MIT Press eBook Chapters

      This paper describes a system for detecting and estimating the properties of multiple travel lanes in an urban road network from calibrated video imagery and laser range data acquired by a moving vehicle. The system operates in several stages on multiple processors, fusing detected road markings, obstacles, and curbs into a stable non-parametric estimate of nearby travel lanes. The system incorporates elements of a provided piecewise-linear road network as a weak prior.Our method is notable in several respects: it estimates multiple travel lanes; it fuses asynchronous, heterogeneous sensor streams; it handles high-curvature roads; and it makes no assumption about the position or orientation of the vehicle with respect to the road.We analyze the system's performance in the context of the 2007 DARPA Urban Challenge. With five cameras and thirteen lidars, it was incorporated into a closed-loop controller to successfully guide an autonomous vehicle through a 90 km urban course at speeds up to 40 km/h amidst moving traffic. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Laser and Vision Based Outdoor Object Mapping

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 9 - 16
      Copyright Year: 2009

      MIT Press eBook Chapters

      Generating rich representations of environments can significantly improve the autonomy of mobile robotics. In this paper we introduce a novel approach to building object-type maps of outdoor environments. Our approach uses conditional random fields (CRF) to jointly classify laser returns in a 2D scan map into seven object types (car, wall, tree trunk, foliage, person, grass, and other). The spatial connectivity of the CRF model is determined via Delaunay triangulation of the laser map. Our model incorporates laser shape features, visual appearance features, structural information extracted from clusters of laser returns, and visual object detectors trained on image data sets available on the internet. The parameters of the CRF are trained from partially labeled laser and camera data collected by a car moving through an urban environment. Our approach achieves 91% accuracy in classifying objects observed along a 3 kilometer trajectory. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Fast Probabilistic Labeling of City Maps

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 17 - 24
      Copyright Year: 2009

      MIT Press eBook Chapters

      This paper introduces a probabilistic, two-stage classification framework for the semantic annotation of urban maps as provided by a mobile robot. During the first stage, local scene properties are considered using a probabilistic bagof- words classifier. The second stage incorporates contextual information across a given scene via a Markov Random Field (MRF). Our approach is driven by data from an onboard camera and 3D laser scanner and uses a combination of appearancebased and geometric features. By framing the classification exercise probabilistically we are able to execute an informationtheoretic bail-out policy when evaluating appearance-based classconditional likelihoods. This efficiency, combined with low order MRFs resulting from our two-stage approach, allows us to generate scene labels at speeds suitable for online deployment and use. We demonstrate and analyze the performance of our technique on data gathered over almost 17 km of track through a city. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Clustering Sensor Data for Terrain Identification using a Windowless Algorithm

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 25 - 32
      Copyright Year: 2009

      MIT Press eBook Chapters

      In this paper we are interested in autonomous systems that can automatically develop terrain classifiers without human interaction or feedback. A key issue is clustering of sensor data from the same terrain. In this context, we present a novel offline windowless clustering algorithm exploiting time-dependency between samples. In terrain coverage, sets of sensory measurements are returned that are spatially, and hence temporally correlated. Our algorithm works by finding a set of parameter values for a user-specified classifier that minimize a cost function. This cost function is related to change in classifier probability outputs over time. The main advantage over other existing methods is its ability to cluster data for fast-switching systems that either have high process or observation noise, or complex distributions that cannot be properly characterized within the average duration of a state. The algorithm was evaluated using three different classifiers (linear separator, mixture of Gaussians and k-NEAREST NEIGHBOR), over both synthetic data sets and mobile robot contact feedback sensor data, with success. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Distributed Localization of Modular Robot Ensembles

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 33 - 40
      Copyright Year: 2009

      MIT Press eBook Chapters

      Internal localization, the problem of estimating relative pose for each module (part) of a modular robot is a prerequisite for many shape control, locomotion, and actuation algorithms. In this paper, we propose a robust hierarchical approach that uses normalized cut to identify dense subregions with small mutual localization error, then progressively merges those subregions to localize the entire ensemble. Our method works well in both 2D and 3D, and requires neither exact measurements nor rigid inter-module connectors. Most of the computations in our method can be effectively distributed. The result is a robust algorithm that scales to large, non-homogeneous ensembles. We evaluate our algorithm in accurate 2D and 3D simulations of scenarios with up to 10,000 modules. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Controlling Shapes of Ensembles of Robots of Finite Size with Nonholonomic Constraints

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 41 - 48
      Copyright Year: 2009

      MIT Press eBook Chapters

      In this paper we focus on the construction of distributed formation control laws that permit the control of individual mobile ground robots in a formation to a desired distribution with minimal knowledge of the global state. As in previous work, we consider an abstraction of the team that is derived from a shape descriptor of the ensemble and the position and orientation of the ensemble. We consider the control of the abstract state with decentralized control laws which are independent of the number of agents. However, we incorporate an important departure from previous work by explicitly modeling the shape of the robot, the geometric, non-interpenetration constraints and nonholonomic, kinematic constraints. Further, we propose a motion planning technique to plan motions for ensembles of robots and a technique for the splitting and merging of groups and subgroups. We demonstrate the effectiveness of the algorithms on a team of differential drive robots in simulation and on real hardware. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Stochastic Recruitment: A Limited-Feedback Control Policy for Large Ensemble Systems

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 49 - 56
      Copyright Year: 2009

      MIT Press eBook Chapters

      This paper is about stochastic recruitment, a control architecture for centrally controlling the ensemble behavior of many identical agents, in a manner similar to motor recruitment in skeletal muscles. Each agent has a finite set of behaviors, or states, which can be switched based on a broadcast command. By switching randomly between states with a centrally determined probability, it is possible to designate the number of agents in each state. This paper covers stochastic recruitment policies for the case when little or no feedback is available from the system. Feed-forward control policies based on rate equilibria are presented, with an analysis of the performance trade-offs inherent in the problem. Minimal feedback control laws are also discussed, and a policy is presented which minimizes the expected convergence time of the system given only the ability to halt the system when the desired output has been achieved. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Prior Data and Kernel Conditional Random Fields for Obstacle Detection

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 57 - 64
      Copyright Year: 2009

      MIT Press eBook Chapters

      We consider the task of training an obstacle detection (OD) system based on a monocular color camera using minimal supervision. We train it to match the performance of a system that uses a laser rangefinder to estimate the presence of obstacles by size and shape. However, the lack of range data in the image cannot be compensated by the extraction of local features alone. Thus, we investigate contextual techniques based on Conditional Random Fields (CRFs) that can exploit the global context of the image, and we compare them to a conventional learning approach. Furthermore, we describe a procedure for introducing prior data in the OD system to increase its performance in “familiar” terrains. Finally, we perform experiments using sequences of images taken from a vehicle for autonomous vehicle navigation applications. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      SARSOP: Efficient Point-Based POMDP Planning by Approximating Optimally Reachable Belief Spaces

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 65 - 72
      Copyright Year: 2009

      MIT Press eBook Chapters

      Motion planning in uncertain and dynamic environments is an essential capability for autonomous robots. Partially observable Markov decision processes (POMDPs) provide a principled mathematical framework for solving such problems, but they are often avoided in robotics due to high computational complexity. Our goal is to create practical POMDP algorithms and software for common robotic tasks. To this end, we have developed a new point-based POMDP algorithm that exploits the notion of optimally reachable belief spaces to improve computational efficiency. In simulation, we successfully applied the algorithm to a set of common robotic tasks, including instances of coastal navigation, grasping, mobile robot exploration, and target tracking, all modeled as POMDPs with a large number of states. In most of the instances studied, our algorithm substantially outperformed one of the fastest existing point-based algorithms. A software package implementing our algorithm is available for download at http://motion.comp.nus.edu. sg/projects/pomdp/pomdp.html. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Detection of Principal Directions in Unknown Environments for Autonomous Navigation

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 73 - 80
      Copyright Year: 2009

      MIT Press eBook Chapters

      Autonomous navigation in unknown but wellstructured environments (e.g., parking lots) is a common task for human drivers and an important goal for autonomous vehicles. In such environments, the vehicles must obey the standard conventions of driving (e.g., passing oncoming vehicles on the correct side), but often lack a map that can be used to guide path planning in an appropriate way. The robots must therefore rely on features of the environment to drive in a safe and predictable way. In this work, we focus on detecting one of such features, the principal directions of the environment.We propose a Markov-random-field (MRF) model for estimating the maximum-likelihood field of principal directions, given the local linear features extracted from the vehicle's sensor data, and show that the method leads to robust estimates of principal directions in complex real-life driving environments. We also demonstrate how the computed principal directions can be used to guide a path-planning algorithm, leading to the generation of significantly improved trajectories. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Approximation Schemes for Two-Player Pursuit Evasion Games with Visibility Constraints

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 81 - 88
      Copyright Year: 2009

      MIT Press eBook Chapters

      In this paper, we consider the problem in which a mobile pursuer attempts to maintain visual contact with an evader as it moves through an environment containing obstacles. This surveillance problem is a variation of traditional pursuitevasion games, with the additional condition that the pursuer immediately loses the game if at any time it loses sight of the evader. We present schemes to approximate the set of initial positions of the pursuer from which it might be able to track the evader.We first consider the case of an environment containing only polygonal obstacles. We prove that in this case the set of initial pursuer configurations from which it does not lose the game is bounded. Moreover, we provide polynomial time approximation schemes to bound this set. We then extend our results to the case of arbitrary obstacles with smooth boundaries. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Numerically Robust LCP Solver for Simulating Articulated Rigid Bodies in Contact

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 89 - 96
      Copyright Year: 2009

      MIT Press eBook Chapters

      This paper presents a numerically robust algorithm for solving linear complementarity problems (LCPs), and applies it to simulation of frictional contacts of articulated rigid bodies each modeled as a general polygonal object.We first point out two problems of the popular pivot-based LCP solver called Lemke Algorithm and its extension with lexicographic ordering, due to numerical errors especially for ill-conditioned LCPs. Our new algorithm solves these problems by storing all pivot candidates and searching for a sequence of pivots that leads to a solution. An LCP-based contact dynamics formulation is combined with a forward dynamics algorithm for articulated rigid bodies to perform the whole simulation using a dynamic programming approach. Simulation examples using a humanoid robot show that the Lemke Algorithm (with or without lexicographic ordering) cannot solve complex contact problems, while our algorithm can successfully simulate such situations. We also demonstrate that the simulation results are qualitatively similar to those of hardware experiments. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Hybrid Motion Planning Using Minkowski Sums

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 97 - 104
      Copyright Year: 2009

      MIT Press eBook Chapters

      Probabilistic and deterministic planners are two major approximate-based frameworks for solving motion planning problems. Both approaches have their own advantages and disadvantages. In this work, we provide an investigation to the following question: Is there a planner that can take the advantages from both probabilistic and deterministic planners? Our strategy to achieve this goal is to use the point-based Minkowski sum of the robot and the obstacles in workspace. Our experimental results show that our new method, called M-sum planner, which uses the geometric properties of Minkowski sum to solve motion planning problems, provides advantages over the existing probabilistic or deterministic planners. In particular, Msum planner is significantly more efficient than the Probabilistic Roadmap Methods (PRMs) and its variants for problems that can be solved by reusing configurations. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Bridging the Gap of Abstraction for Probabilistic Decision Making on a Multi-Modal Service Robot

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 105 - 110
      Copyright Year: 2009

      MIT Press eBook Chapters

      This paper proposes a decision making and control supervision system for a multi-modal service robot. With partially observable Markov decision processes (POMDPs) utilized for scenario level decision making, the robot is able to deal with uncertainty in both observation and environment dynamics and can balance multiple, conflicting goals. By using a flexible task sequencing system for fine grained robot component coordination, complex sub-activities, beyond the scope of current POMDP solutions, can be performed. The sequencer bridges the gap of abstraction between abstract POMDP models and the physical world concerning actions, and in the other direction multi-modal perception is filtered while preserving measurement uncertainty and model-soundness. A realistic scenario for an autonomous, anthropomorphic service robot, including the modalities of mobility, multi-modal humanrobot interaction and object grasping, has been performed robustly by the system for several hours. The proposed filter- POMDP reasoner is compared with classic POMDP as well as MDP decision making and a baseline finite state machine controller on the physical service robot, and the experiments exhibit the characteristics of the different algorithms. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Adaptive Body Scheme Models for Robust Robotic Manipulation

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 111 - 118
      Copyright Year: 2009

      MIT Press eBook Chapters

      Truly autonomous systems require the ability to monitor and adapt their internal body scheme throughout their entire lifetime. In this paper, we present an approach allowing a robot to learn from scratch and maintain a generative model of its own physical body through self-observation with a single monocular camera. We represent the robot's internal model as a compact Bayesian network, consisting of local models that describe the physical relationships between neighboring body parts. We introduce a flexible Bayesian framework that allows to simultaneously select the maximum-likely network structure and to learn the underlying conditional density functions. Changes in the robot's physiology can be detected by identifying mismatches between model predictions and the self-perception. To quickly adapt the model to changed situations, we developed an efficient search heuristic that starts from the structure of the best explaining memorized network and then replaces local components where necessary. In experiments carried out with a real robot equipped with a 6-DOF manipulator as well as in simulation, we show that our system can quickly adapt to changes of the body physiology in full 3D space, in particular with limited visibility, noisy and partially missing observations, and without the need for proprioception. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Bearing-Only Control Laws For Balanced Circular Formations of Ground Robots

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 119 - 126
      Copyright Year: 2009

      MIT Press eBook Chapters

      For a group of constant-speed ground robots, a simple control law is designed to stabilize the motion of the group into a balanced circular formation using a consensus approach. It is shown that the measurements of the bearing angles between the robots are sufficient for reaching a balanced circular formation. We consider two different scenarios that the connectivity graph of the system is either a complete graph or a ring. Collision avoidance capabilities are added to the team members and the effectiveness of the control laws are demonstrated on a group of mobile robots. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      CPG-based Control of a Turtle-like Underwater Vehicle

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 127 - 134
      Copyright Year: 2009

      MIT Press eBook Chapters

      We present a new bio-inspired control strategy for an autonomous underwater vehicle by constructing coupled nonlinear oscillators, similar to the animal central pattern generators (CPGs). Using contraction theory, we show that the network of oscillators globally converges to a specific pattern of oscillation. We experimentally validate the proposed control law using a turtle-like underwater vehicle, whose fin actuators successfully exhibit a pattern that resembles the motion of fore limbs of a swimming sea turtle. In order to further fulfill the potential of the CPG-based control, we propose to feed back the actuator states to the coupled oscillators, thereby achieving not only the synchronization of the oscillators, but also the synchronization of actual foil states. Such a closed-loop version of CPGs makes the controller more robust and practical in the presence of external disturbances. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      HyPE: Hybrid Particle-Element Approach for Recursive Bayesian Searching-and-Tracking

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 135 - 142
      Copyright Year: 2009

      MIT Press eBook Chapters

      This paper presents a hybrid particle-element approach, HyPE, suitable for recursive Bayesian searching-andtracking (SAT). The hybrid concept, to synthesize two recursive Bayesian estimation (RBE) methods to represent and maintain the belief about all states in a dynamic system, is distinct from the concept behind “mixed approaches“, such as Rao-Blackwellized particle filtering, which use different RBE methods for different states. HyPE eliminates the need for computationally expensive numerical integration in the prediction stage and allows space reconfiguration, via remeshing, at minimal computational cost. Numerical examples show the efficacy of the hybrid approach, and demonstrate its superior performance in SAT scenarios when compared with both the particle filter and the element-based method. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Abstractions and Algorithms for Cooperative multiple Robot Planar Manipulation

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 143 - 150
      Copyright Year: 2009

      MIT Press eBook Chapters

      In this paper, we will study abstractions and algorithms for planar manipulation systems using two cooperating robots under uncertainties. We propose a formal framework for developing abstractions, which are simpler models of the original systems that preserve properties of interest to facilitate the development of planning and control algorithms. Our abstractions are derived from robust motion primitives that correspond to control inputs leading to system trajectories which preserve the properties of interest under uncertainties. We then use the proposed framework to construct an abstraction and design planning and control algorithms for a multiple robot cooperative manipulation system. Finally, we present experimental results to validate our approach. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      A Local Collision Avoidance Method for Non-strictly Convex Polyhedra

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 151 - 158
      Copyright Year: 2009

      MIT Press eBook Chapters

      This paper proposes a local collision avoidance method for non-strictly convex polyhedra with continuous velocities. The main contribution of the method is that non-strictly convex polyhedra can be used as geometric models of the robot and the environment without any approximation. The problem of the continuous interaction generation between polyhedra is reduced to the continuous constraints generation between polygonal faces and the continuity of those constraints are managed by the combinatorics based on Voronoi regions of a face. A collision-free motion is obtained by solving an optimization problem defined by an objective function which describes a task and linear inequality constraints which do geometrical constraints to avoid collisions. The proposed method is examined using example cases of simple objects and also applied to a humanoid robot HRP-2. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      BiSpace Planning: Concurrent Multi-Space Exploration

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 159 - 166
      Copyright Year: 2009

      MIT Press eBook Chapters

      We present a planning algorithm called BiSpace that produces fast plans to complex high-dimensional problems by simultaneously exploring multiple spaces. We specifically focus on finding robust solutions to manipulation and grasp planning problems by using BiSpace's special characteristics to explore the work and configuration spaces of the environment and robot. Furthermore, we present a number of techniques for constructing informed heuristics to intelligently search through these highdimensional spaces. In general, the BiSpace planner is applicable to any problem involving multiple search spaces. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Structural Improvement Filtering Strategy for PRM

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 167 - 174
      Copyright Year: 2009

      MIT Press eBook Chapters

      Sampling based motion planning methods have been highly successful in solving many high degree of freedom motion planning problems arising in diverse application domains such as traditional robotics, computer-aided design, and computational biology and chemistry. Recent work in metrics for sampling based planners provide tools to analyze the model building process at three levels of detail: sample level, region level, and global level. These tools are useful for comparing the evolution of sampling methods, and have shown promise to improve the process altogether [15], [17], [24].Here, we introduce a filtering strategy for the Probabilistic Roadmap Methods (PRM) with the aim to improve roadmap construction performance by selecting only the samples that are likely to produce roadmap structure improvement. By measuring a new sample's maximum potential structural improvement with respect to the current roadmap, we can choose to only accept samples that have an adequate potential for improvement. We show how this approach can improve the standard PRM framework in a variety of motion planning situations using popular sampling techniques. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Model Based Vehicle Tracking for Autonomous Driving in Urban Environments

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 175 - 182
      Copyright Year: 2009

      MIT Press eBook Chapters

      Situational awareness is crucial for autonomous driving in urban environments. This paper describes moving vehicle tracking module that we developed for our autonomous driving robot Junior. The robot won second place in the Urban Grand Challenge, an autonomous driving race organized by the U.S. Government in 2007. The tracking module provides reliable tracking of moving vehicles from a high-speed moving platform using laser range finders. Our approach models both dynamic and geometric properties of the tracked vehicles and estimates them using a single Bayes filter per vehicle. We also show how to build efficient 2D representations out of 3D range data and how to detect poorly visible black vehicles. Experimental validation includes the most challenging conditions presented at the UGC as well as other urban settings. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Improving Localization Robustness in Monocular SLAM Using a High-Speed Camera

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 183 - 190
      Copyright Year: 2009

      MIT Press eBook Chapters

      In the robotics community localization and mapping of an unknown environment is a well-studied problem. To solve this problem in real-time using visual input, a standard monocular Simultaneous Localization and Mapping (SLAM) algorithm can be used. This algorithm is very stable when smooth motion is expected, but in case of erratic or sudden movements, the camera pose typically gets lost. To improve robustness in Monocular SLAM (MonoSLAM) we propose to use a camera with faster readout speed to obtain a frame rate of 200Hz. We further present an extended MonoSLAM motion model, which can handle movements with significant jitter. In this work the improved localization and mapping have been evaluated against ground truth, which is reconstructed from off-line vision. To explain the benefits of using a high frame rate vision input in MonoSLAM framework, we performed repeatable experiments with a high-speed camera mounted onto a robotic arm. Due to the dense visual information MonoSLAM can faster shrink localization and mapping uncertainties and can operate under fast, erratic, or sudden movements. The extended motion model can provide additional robustness against significant handheld jitter when throwing or shaking the camera. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Simplex-Tree Based Kinematics of Foldable Objects as Multi-body Systems Involving Loops

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 191 - 198
      Copyright Year: 2009

      MIT Press eBook Chapters

      Many practical multi-body systems involve loops. Studying the kinematics of such systems has been challenging, partly because of the requirement of maintaining loop closure constraints, which have conventionally been formulated as highly nonlinear equations in joint parameters. Recently, novel parameters defined by trees of triangles have been introduced for a broad class of linkage systems involving loops (e.g., spatial loops with spherical joints and planar loops with revolute joints); these parameters greatly simplify kinematics related computations and endow system configuration spaces with highly tractable piecewise convex geometries. In this paper, we describe a more general approach for multi-body systems, with loops, that allow construction trees of simplices. We illustrate the applicability and efficiency of our simplex-tree based approach to kinematics by a study of foldable objects. We present two sets of new parameters for single-vertex rigid fold kinematics; like the parameters in the triangle-tree prototype, each has a geometrically meaningful and computationally tractable constraint formulation, and each endows the configuration space with a nice geometry. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      fMRI-Compatible Robotic Interfaces with Fluidic Actuation

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 199 - 205
      Copyright Year: 2009

      MIT Press eBook Chapters

      Actuation is a major challenge in the development of robotic systems intended to work in functional Magnetic Resonance Imaging (fMRI) procedures, due to the high magnetic fields and limited space in the scanner. Fluidic actuators can be made fMRI-compatible and are, thus, promising solutions. In this work we developed two robotic interface devices, one with hydraulic and another with pneumatic actuation, to control one degree-of-freedom translational movements of a user that performs fMRI tasks. Due to the fMRI-compatibility restrictions, special materials were used for the endeffector which works in the scanner bore, and active components such as the control valves and pressure sensors, had to be placed far away from the endeffector with long transmission lines in between. Therefore, the two fMRI-compatible setups differed from conventional fluidic actuation systems and brought control difficulties. Both systems have been proved to be fMRI-compatible and yield no image artifacts in a 3T scanner. Passive as well as active subject movements were realized by classical position and impedance controllers. With the hydraulic system we achieved smoother movements, higher position control accuracy and improved robustness against force disturbances than with the pneumatic system. In contrast, the pneumatic system was back-drivable, showed faster dynamics with relatively low pressure, and allowed force control. Furthermore, it is easier to maintain and does not cause hygienic problems after leakages. In general, pneumatic actuation is favorable for fast or force-controlled applications, whereas hydraulic actuation is recommended for applications that require high position accuracy, or slow and smooth movements. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Proofs and Experiments in Scalable, Near-Optimal Search by Multiple Robots

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 206 - 213
      Copyright Year: 2009

      MIT Press eBook Chapters

      In this paper, we examine the problem of locating a non-adversarial target using multiple robotic searchers. This problem is relevant to many applications in robotics including emergency response and aerial surveillance. Assuming a known environment, this problem becomes one of choosing searcher paths that are most likely to intersect with the path taken by the target. We refer to this as the Multi-robot Efficient Search Path Planning (MESPP) problem. Such path planning problems are NP-hard, and optimal solutions typically scale exponentially in the number of searchers. We present a finitehorizon path enumeration algorithm for solving the MESPP problem that utilizes sequential allocation to achieve linear scalability in the number of searchers. We show that solving the MESPP problem requires the maximization of a nondecreasing, submodular objective function, which directly leads to theoretical guarantees on paths generated by sequential allocation. We also demonstrate how our algorithm can run online to incorporate noisy measurements of the target's position during search. We verify the performance of our algorithm both in simulation and in experiments with a novel radio sensor capable of providing range through walls. Our results show that our linearly scalable MESPP algorithm generates searcher paths competitive with those generated by exponential algorithms. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Planning Long Dynamically-Feasible Maneuvers for Autonomous Vehicles

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 214 - 221
      Copyright Year: 2009

      MIT Press eBook Chapters

      In this paper, we present an algorithm for generating complex dynamically-feasible maneuvers for autonomous vehicles traveling at high speeds over large distances. Our approach is based on performing anytime incremental search on a multiresolution, dynamically-feasible lattice state space. The resulting planner provides real-time performance and guarantees on and control of the suboptimality of its solution. We provide theoretical properties and experimental results from an implementation on an autonomous passenger vehicle that competed in, and won, the Urban Challenge competition. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Friction-Induced Velocity Fields for Point Parts Sliding on a Rigid Oscillated Plate

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 222 - 229
      Copyright Year: 2009

      MIT Press eBook Chapters

      We show that small-amplitude periodic motion of a rigid plate causes point parts on the plate to move as if they are in a position-dependent velocity field. Further, we prove that every periodic plate motion maps to a unique velocity field. By allowing a plate to oscillate with six-degrees-of-freedom, we can create a large family of programmable velocity fields.We examine in detail sinusoidal plate motions that generate fields with either isolated sinks or squeeze lines. These fields can be exploited to perform tasks such as sensorless part orientation. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Metastable Walking on Stochastically Rough Terrain

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 230 - 237
      Copyright Year: 2009

      MIT Press eBook Chapters

      Simplified models of limit-cycle walking on flat terrain have provided important insights into the nature of legged locomotion. Real walking robots (and humans), however, do not exhibit true limit cycle dynamics because terrain, even in a carefully designed laboratory setting, is inevitably non-flat. Walking systems on stochastically rough terrain may not satisfy strict conditions for limit-cycle stability but can still demonstrate impressively long-living periods of continuous walking. Here, we examine the dynamics of rimless-wheel and compass-gait walking on randomly generated rough terrain and employ tools from stochastic processes to describe the ‘stochastic stability’ of these gaits. This analysis generalizes our understanding of walking stability and may provide statistical tools for experimental limit cycle analysis on real walking systems. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Target Enumeration via Integration Over Planar Sensor Networks

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 238 - 245
      Copyright Year: 2009

      MIT Press eBook Chapters

      We solve the problem of counting the total number of observable targets (e.g., persons, vehicles, etc.) in a region based on local counts performed by sensors which measure only the number of targets nearby and neither their identities nor any positional information. This theory is robust and accommodates ad hoc sensor networks and mobile robot sensors alike. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Using Recognition to Guide a Robot's Attention

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 246 - 253
      Copyright Year: 2009

      MIT Press eBook Chapters

      In the transition from industrial to service robotics, robots will have to deal with increasingly unpredictable and variable environments. We present a system that is able to recognize objects of a certain class in an image and to identify their parts for potential interactions. This is demonstrated for object instances that have never been observed during training, and under partial occlusion and against cluttered backgrounds. Our approach builds on the Implicit Shape Model of Leibe and Schiele, and extends it to couple recognition to the provision of meta-data useful for a task. Meta-data can for example consist of part labels or depth estimates. We present experimental results on wheelchairs and cars. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Learning to Manipulate Articulated Objects in Unstructured Environments Using a Grounded Relational Representation

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 254 - 261
      Copyright Year: 2009

      MIT Press eBook Chapters

      We introduce a learning-based approach to manipulation in unstructured environments. This approach permits autonomous acquisition of manipulation expertise from interactions with the environment. The resulting expertise enables a robot to perform effective manipulation based on partial state information. The manipulation expertise is represented in a relational state representation and learned using relational reinforcement learning. The relational representation renders learning tractable by collapsing a large number of states onto a single, relational state. The relational state representation is carefully grounded in the perceptual and interaction skills of the robot. This ensures that symbolically learned knowledge remains meaningful in the physical world. We experimentally validate the proposed learning approach on the task of manipulating an articulated object to obtain a model of its kinematic structure. Our experiments demonstrate that the manipulation expertise acquired by the robot leads to substantial performance improvements. These improvements are maintained when experience is applied to previously unseen objects. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      High Performance Outdoor Navigation from Overhead Data using Imitation Learning

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 262 - 269
      Copyright Year: 2009

      MIT Press eBook Chapters

      High performance, long-distance autonomous navigation is a central problem for field robotics. Efficient navigation relies not only upon intelligent onboard systems for perception and planning, but also the effective use of prior maps and knowledge. While the availability and quality of low cost, high resolution satellite and aerial terrain data continues to rapidly improve, automated interpretation appropriate for robot planning and navigation remains difficult. Recently, a class of machine learning techniques have been developed that rely upon expert human demonstration to develop a function mapping overhead data to traversal cost. These algorithms choose the cost function so that planner behavior mimics an expert's demonstration as closely as possible. In this work, we extend these methods to automate interpretation of overhead data. We address key challenges, including interpolation-based planners, non-linear approximation techniques, and imperfect expert demonstration, necessary to apply these methods for learning to search for effective terrain interpretations. We validate our approach on a large scale outdoor robot during over 300 kilometers of autonomous traversal through complex natural environments. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Classifying Dynamic Objects: An Unsupervised Learning Approach

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 270 - 277
      Copyright Year: 2009

      MIT Press eBook Chapters

      For robots operating in real-world environments, the ability to deal with dynamic entities such as humans, animals, vehicles, or other robots is of fundamental importance. The variability of dynamic objects, however, is large in general, which makes it hard to manually design suitable models for their appearance and dynamics. In this paper, we present an unsupervised learning approach to this model-building problem. We describe an exemplar-based model for representing the time-varying appearance of objects in planar laser scans as well as a clustering procedure that builds a set of object classes from given training sequences. Extensive experiments in real environments demonstrate that our system is able to autonomously learn useful models for, e.g., pedestrians, skaters, or cyclists without being provided with external class information. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Probabilistic Models of Object Geometry for Grasp Planning

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 278 - 285
      Copyright Year: 2009

      MIT Press eBook Chapters

      Robot manipulators generally rely on complete knowledge of object geometry in order to plan motions and compute successful grasps. However, manipulating real-world objects poses a substantial modelling challenge. New instances of known object classes may vary from learned models. Objects that are not perfectly rigid may appear in new configurations that do not match any of the known geometries.In this paper we describe an algorithm for learning generative probabilistic models of object geometry for the purposes of manipulation; these models capture both non-rigid deformations of known objects and variability of objects within a known class. Given a single image of partially occluded objects, the model can be used to recognize objects based on the visible portion of each object contour, and then estimate the complete geometry of the object to allow grasp planning.We provide two main contributions: a probabilistic model of shape geometry and a graphical model for performing correspondence between shape descriptions. We show examples of learned models from image data and demonstrate how the learned models can be used by a manipulation planner to grasp objects in cluttered visual scenes. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Dynamic Modeling of Stick Slip Motion in an Untethered Magnetic Micro-Robot

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 286 - 293
      Copyright Year: 2009

      MIT Press eBook Chapters

      This work presents the dynamic modeling of an untethered electromagnetically actuated magnetic micro-robot, and compares computer simulations to experimental results. The micro-robot, which is composed of neodymium-iron-boron with dimensions 250 µm x 130 µm x 100 µm, is actuated by a system of 5 macro-scale electromagnets. Periodic magnetic fields are created using two different control methods, which induce stick-slip motion in the micro-robot. The effects of model parameter variations on micro-robot velocity are explored and discussed. Micro-robot stick-slip motion is accurately captured in simulation. Velocity trends of the micro-robot on a silicon surface as a function of magnetic field oscillation frequency and magnetic field strength are also captured. Mismatch between simulation and reality is discussed. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Super-Flexible Skin Sensors Embedded on the Whole Body, Self-Organizing Based on Haptic Interactions

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 294 - 301
      Copyright Year: 2009

      MIT Press eBook Chapters

      As robots become more ubiquitous in our daily lives, humans and robots are working in ever-closer physical proximity to each other. These close physical distances change the nature of human robot interaction considerably. First, it becomes more important to consider safety, in case robots accidentally touch (or hit) the humans. Second, touch (or haptic) feedback from humans can be a useful additional channel for communication, and is a particularly natural one for humans to utilize. Covering the whole robot body with malleable tactile sensors can help to address the safety issues while providing a new communication interface. First, soft, compliant surfaces are less dangerous in the event of accidental human contact. Second, flexible sensors are capable of distinguishing many different types of touch (e.g., hard v.s. gentle stroking). Since soft skin on a robot tends to invite humans to engage in even more touch interactions, it is doubly important that the robot can process haptic feedback from humans. In this paper, we discuss attempts to solve some of the difficult new technical and information processing challenges presented by flexible touch sensitive skin. Our approach is based on a method for sensors to self-organize into sensor banks for classification of touch interactions. This is useful for distributed processing and helps to reduce the maintenance problems of manually configuring large numbers of sensors. We found that using sparse sensor banks containing as little as 15% of the full sensor set it is possible to classify interaction scenarios with accuracy up to 80% in a 15-way forced choice task. Visualization of the learned subspaces shows that, for many categories of touch interactions, the learned sensor banks are composed mainly of physically local sensor groups. These results are promising and suggest that our proposed method can be effectively used for automatic analysis of touch behaviors in more complex tasks. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      NanoNewton Force Sensing and Control in Microrobotic Cell Manipulation

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 302 - 309
      Copyright Year: 2009

      MIT Press eBook Chapters

      Cellular force sensing and control techniques are capable of enhancing the dexterity and reliability of microrobotic cell manipulation systems. This paper presents a vision-based cellular force sensing technique using a microfabricated elastic cell holding device and a sub-pixel visual tracking algorithm for resolving forces down to 3.7nN during microrobotic mouse embryo injection. The technique also experimentally proves useful for in situ differentiation of healthy mouse embryos from those with compromised developmental competence without the requirement of a separate mechanical characterization process. Concerning force-controlled microrobotic cell manipulation (pick-transport-place), this paper presents the first demonstration of nanoNewton force-controlled cell micrograsping using a MEMS-based microgripper with integrated two-axis force feedback. On-chip force sensors are used for detecting contact between the microgripper and cells to be manipulated (resolution: 38.5nN) and sensing gripping forces (resolution: 19.9nN) during force-controlled grasping. The experimental results demonstrate that the microgripper and the control system are capable of rapid contact detection and reliable force-controlled micrograsping to accommodate variations in size and stiffness of cells with a high reproducibility. View full abstract»

    • Full text access may be available. Click article title to sign in or learn about subscription options.

      Gas Distribution Modeling using Sparse Gaussian Process Mixture Models

      Brock, O. ; Trinkle, J. ; Ramos, F.
      Robotics:Science and Systems IV

      Page(s): 310 - 317
      Copyright Year: 2009

      MIT Press eBook Chapters

      In this paper, we consider the problem of learning a two dimensional spatial model of a gas distribution with a mobile robot. Building maps that can be used to accurately predict the gas concentration at query locations is a challenging task due to the chaotic nature of gas dispersal. We present an approach that formulates this task as a regression problem. To deal with the specific properties of typical gas distributions, we propose a sparse Gaussian process mixture model. This allows us to accurately represent the smooth background signal as well as areas of high concentration. We integrate the sparsification of the training data into an EM procedure used for learning the mixture components and the gating function. Our approach has been implemented and tested using datasets recorded with a real mobile robot equipped with an electronic nose. We demonstrate that our models are well suited for predicting gas concentrations at new query locations and that they outperform alternative methods used in robotics to carry out in this task. Index Terms—Gas distribution modeling, gas sensing, Gaussian processes, mixture models View full abstract»




| Create Account

IEEE Account

Purchase Details

Profile Information

Need Help?


IEEE Advancing Technology for Humanity About IEEE Xplore | Contact | Help | Terms of Use | Nondiscrimination Policy | Site Map | Privacy & Opting Out of Cookies

A not-for-profit organization, IEEE is the world's largest professional association for the advancement of technology.
© Copyright 2014 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.