Reinforcement Learning-based Motion Generation for a Tracked Robot to Go Over a Sphere-shaped Non-fixed Obstacle.
Hidenori TakamiyaRyosuke YajimaJun Younes Louhi KasaharaRen KomatsuKeiji NagataniAtsushi YamashitaHajime AsamaPublished in: SII (2023)
Keyphrases
- mobile robot
- reinforcement learning
- motion control
- humanoid robot
- image sequences
- motion planning
- real robot
- robot control
- autonomous navigation
- robot motion
- end effector
- real time
- configuration space
- autonomous robots
- camera motion
- control signals
- vision system
- position and orientation
- motion model
- path planning
- motion analysis
- monocular vision
- perceptual aliasing
- robot arm
- robot moves
- sagittal plane
- multi robot
- optical flow
- space time
- machine learning
- robotic systems
- human robot interaction
- motion segmentation
- visual servoing
- human motion
- parallel robot
- inverse kinematics
- motion estimation
- function approximation
- motion patterns
- rotational motion
- computer vision
- robotic tasks
- motion detection
- autonomous learning
- video sequences
- degrees of freedom
- control system
- moving objects
- state space
- robot navigation