Synergistic Task and Motion Planning With Reinforcement Learning-Based Non-Prehensile Actions.
Gaoyuan LiuJoris De WinterDenis SteckelmacherRoshan Kumar HotaAnn NowéBram VanderborghtPublished in: IEEE Robotics Autom. Lett. (2023)
Keyphrases
- motion planning
- reinforcement learning
- degrees of freedom
- action selection
- perceptual aliasing
- trajectory planning
- path planning
- mobile robot
- partially observable
- robot arm
- humanoid robot
- robotic arm
- action space
- function approximation
- state space
- reward function
- robotic tasks
- obstacle avoidance
- multi robot
- learning algorithm
- inverse kinematics
- learning agent
- belief space
- manipulation tasks
- autonomous mobile robot
- mechanical systems
- configuration space
- real time
- climbing robot
- model free
- decision theoretic
- machine learning
- robot control
- human activities
- decision problems
- markov decision processes
- kinematic model
- pose estimation
- multi agent