Train a robot arm with reinforcement learning in AMD Schola using Unreal® Engine, progressively increasing task complexity to adapt to changing conditions.

AMD GPUOpen

Demonstrates training a UFACTORY X-ARM 5 robotic arm using reinforcement learning with AMD Schola v2.0 and Unreal Engine. The tutorial progresses through three increasingly complex tasks: reaching a fixed target, reaching randomized target positions, and handling both random targets and locations. Uses the SAC algorithm from Stable-Baselines3 with continuous observation and action spaces, showing how to build robust models that adapt to dynamic conditions through progressive complexity.

Training an X-ARM 5 robotic arm with AMD Schola and Unreal Engine