Home▸Robotics & Kinematics▸Robot Parkour (2D)

🤖 Robot Parkour (2D)

A top-down Canvas2D reinforcement-learning training view: robot agents learn to navigate an obstacle course toward goal beacons.

Robotics & Kinematics2DModerate60 FPS📱 Mobile-adapted⇄ 3D version

⚙ Under the hood

Top-down Canvas2D reinforcement-learning training view: a fleet of robot agents learns to weave through scattered obstacles toward goal beacons, with live reward, loss and entropy telemetry as the policy converges.

RoboticsReinforcement LearningPathfindingAI TrainingTop-down

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install