Bounded Energy Collisions in Human–Robot Cooperative Transportation

Alevizos, Konstantinos; Bechlioulis, Charalampos P.; Kyriakopoulos, Kostas J.

doi:10.1109/tmech.2022.3150247

Cited by 4 publications

(2 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other tasks such as formation, flocking are also studied under various constraints, see [18]. Another relevant task is the collaborative load transportation [19], [20], where several agents transport one object to the destination via pushing or grasping. These motion planning problems remain challenging due to the dynamic and geometric constraints associated with different tasks.…”

Section: A Related Workmentioning

confidence: 99%

“…For simplicity, each robot has a first-order dynamics and can provide bounded pushing force at specific points of the box, thus changing the state of boxes with second-order translational invariant dynamics. Note that the cluttered workspace introduces severe geometric constraints, yielding a significantly more difficult transportation task than the obstacle-free environments [19], [20]. Initially, the task cost f p¨q in ( 7) is estimated by the sum of Euclidean distances from agents to the targeted box, and the distance to its goal position.…”

Section: Numerical Experimentsmentioning

confidence: 99%

See 1 more Smart Citation

Constraint-Based Multi-Agent Reinforcement Learning for Collaborative Tasks

Shang

Karamouzas

et al. 2023

Preprint

View full text Add to dashboard Cite

In order to be successfully executed, collaborative tasks performed by two agents often require a cooperative strategy to be learned. In this work, we propose a constraint-based multi-agent reinforcement learning approach called Constrained Multi-agent Soft Actor Critic (C-MSAC) to train control policies for simulated agents performing collaborative multi-phase tasks. Given a task with n phases, the first n-1 phases are treated as constraints for the final task phase objective, which is addressed with a centralized training and decentralized execution approach. We highlight our framework on a tray balancing task including two phases: tray lifting and cooperative tray control for target following. We evaluate our proposed approach and compare it against its unconstrained variant (MSAC). The performed comparisons show that C-MSAC leads to higher success rates, more robust control policies, and better generalization performance.

show abstract

Section: A Related Workmentioning

confidence: 99%

Section: Numerical Experimentsmentioning

confidence: 99%