multi-arm

Design of single- and multi-stage multi-arm clinical trials

Reinforcement learning environment for UR5e arms with MuJoCo 3 — SAC training for reach, pick-and-place, and symmetric multi-arm cooperative tasks. Includes a ROS 2 policy node for Gazebo deployment.

SimTOST is an R package designed to facilitate the sample size estimation for bioequivalence studies using a simulation-based approach.

Variety of Multi-Arm Bandit (MAB) algorithms using classic and advanced strategies, including tools for experiments and simulations in stationary and nonstationary environments

MuJoCo and robosuite platform for multi-arm teleoperation, demonstration collection, replay, and VLA dataset preparation