3

Model-Informed Safe Reinforcement Learning for Bipedal Locomotion via Step-to-Step Prediction
Reduced-model safety certificates inform reinforcement learning and runtime foot-placement filtering in Digit simulation.
PDF Cite Project Project Video arXiv
Perceptive Variable-Timing Footstep Planning for Humanoid Locomotion on Disconnected Footholds
Depth-based steppable regions and variable-timing planning evaluated in Digit simulation.
PDF Cite Project Video arXiv
Reinforcement learning with data bootstrapping for dynamic subgoal pursuit in humanoid robot navigation
Data-bootstrapped reinforcement learning selects navigation subgoals for an MPC gait planner, evaluated in Digit simulation.
PDF Cite Project Project Video arXiv
Unified path and gait planning for safe bipedal robot navigation
Unified path and gait planning evaluated in Digit simulation with randomly generated obstacles.
PDF Cite Project Video arXiv
Unified path and gait planning for safe bipedal robot navigation
Real-Time Navigation for Bipedal Robots in Dynamic Environments
The popularity of mobile robots has been steadily growing, with these robots being increasingly utilized to execute tasks previously …
PDF Cite Project Video arXiv
Real-Time Navigation for Bipedal Robots in Dynamic Environments
Adaptive feedback regulator for powered lower-limb exoskeleton under model uncertainty
This paper presents a neural network (NN) based adaptive feedback regulator to ensure the lateral and longitudinal stability and …
PDF Cite Project arXiv
Adaptive feedback regulator for powered lower-limb exoskeleton under model uncertainty