Collision-free Navigation of a Sensorless Autonomous Robot in Indoor Environments with Moving and Steady Obstacles

This paper presents a novel method to navigate a sensorless autonomous mobile robot in indoor environments. Fixed sensor networks and computing hardware are used to monitor targets and the robot continuously. The fixed infrastructure component localises the mobile robot, simultaneously tracks all dynamic obstacles and generates the robot’s trajectory. The authors have implemented the Interacting Multiple Model Joint Probabilistic Data Association (IMM-JPDA) filter to track multiple targets, including the robot, in cluttered environments. The proposed innovative approach separates the tasks of sensing, trajectory planning and low-level control, where mobile robots only perform low-level control. High-fidelity simulations are performed in MATLAB software to illustrate how the presented solution works.