Integrated Task and Motion Planning for Multi-Robot Manipulation in Industry and Service Automation

This paper introduces a fully integrated task-and-motion planning (TAMP) framework for robotic manipulation in dynamic and uncertain environments, particularly in multi-arm manipulator systems where reachability constraints change over time. Existing approaches, including probabilistic road maps and traditional TAMP frameworks, struggle with high-dimensionality, infeasible actions, and limited adaptability. To address these challenges, we propose a novel planning system that seamlessly integrates high-level symbolic task planning with low-level motion execution via a Dynamic Shared-Space Graph (D-SSG) and a Planning Domain Definition Language (PDDL)-based interface layer. We couple ROSPlan-based symbolic planning with a Dynamic Shared-Space Graph (D-SSG) to enable planning-level real-time adaptation: dynamic validation of actions, local plan repair, and execution updates without full re-planning. A PDDL-driven translation layer links predicates and effects to geometric feasibility checks and motion primitives. In ROS/Gazebo studies with multi-arm setups, the framework improves success rates and reduces planning time in semi-dynamic environments. These results verify the proposed framework as a promising solution for advancing multi-robot coordination in industrial and service automation applications.