Answers are grounded in this paper and cite their sources.
This paper describes a robot agnostic approach to ensure kinematic feasibility of a mobile manipulator. In this work a mobile manipulator’s ability to pick and place tools on a shop floor has been presented, where the two tasks are hindered by accurate movement of the platform and the joint limits of the manipulator combined with less coordination between the platform navigation controller and the manipulation controller. The robotic platform presented in this paper uses two motion planners one for the manipulator and the other for the platform. The planner for the manipulator takes into consideration current position of the platform and its joint limits to plan motion to a 3D location which may be sometimes unreachable. Similarly, if the navigation errors of the platform are quite high the feasibility of the arm to perform pick and place task is significantly reduced. Therefore use of these two independent planner results in sub-optimal solutions. To pro-actively help both the planners this paper describes a Behavior Tree (BT) based forward simulation technique in-order to plan ahead for the task by finding suitable platform stopping position while also re-actively avoiding object detection, pick and place failures by performing dynamic manipulation planning. This paper also validates such forward simulation technique with a mobile manipulator operating in an automotive human robot collaboration scenario.
No takes yet. Share an insight, caveat, or question.
Deshpande et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: