Learning Robot Trajectories subject to Kinematic Joint Constraints

Kiemel, Jonas C.; Kröger, Torsten

doi:10.1109/icra48506.2021.9561159

Cited by 6 publications

(8 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, penalties for undesired behaviors can be added to the reward function of an unconstrained Markov decision process (MDP) [14], [15]. Although penalties reduce the likelihood of undesirable behaviors, safety violations are not entirely prevented, even if the training process is carried out until convergence [6]. In some cases, task-specific heuristics can be used to avoid unsafe behaviors [1].…”

Section: B Learning Safe Motions In Roboticsmentioning

confidence: 99%

“…However, this approach is often very restrictive and not suitable for all types of constraints [9]. Recently, an action space representation to ensure compliance with kinematic joint constraints was proposed [6]. Conflicting constraints are avoided over an infinite time-horizon and the work space of the robot is not restricted.…”

Section: B Learning Safe Motions In Roboticsmentioning

confidence: 99%

“…Conflicting constraints are avoided over an infinite time-horizon and the work space of the robot is not restricted. We utilize the code provided by [6] for our action space representation and to generate braking trajectories. When considering kinematic joint constraints only, all joints can be treated as decoupled.…”

Section: B Learning Safe Motions In Roboticsmentioning

confidence: 99%

“…Our method addresses the problem of learning collisionfree robot motions complying with kinematic and dynamic joint limits. Like in [6], the following kinematic constraints are defined for each revolute robot joint:…”

Section: A Problem Statementmentioning

confidence: 99%

“…• Compliance with kinematic joint limits is ensured by the design of the action space used for reinforcement learning [6]. • Collisions and torque limit violations are prevented by ensuring the existence of an alternative safe behavior at each decision step [7].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Learning Collision-free and Torque-limited Robot Trajectories based on Alternative Safe Behaviors

Kiemel¹,

Kröger²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

This paper presents an approach to learn online generation of collision-free and torque-limited trajectories for industrial robots. A neural network, which is trained via reinforcement learning, is periodically invoked to predict future motions. For each robot joint, the network outputs the kinematic state that is desired at the end of the current time interval. Compliance with kinematic joint limits is ensured by the design of the action space. Given the current kinematic state and the network prediction, a trajectory for the current time interval can be computed. The main idea of our paper is to execute the predicted motion only if a collision-free and torquelimited way to continue the trajectory is known. In practice, the predicted motion is expanded by a braking trajectory and simulated using a physics engine. If the simulated trajectory complies with all safety constraints, the predicted motion is carried out. Otherwise, the braking trajectory calculated in the previous decision step serves as an alternative safe behavior. For evaluation, up to three simulated robots are trained to reach as many randomly placed target points as possible. We show that our method reliably prevents collisions with static obstacles and collisions between the robots, while generating motions that respect both torque limits and kinematic joint limits. Experiments with a real robot demonstrate that safe trajectories can be generated in real-time.

show abstract