Skip to content

Robot Learning

  • Robot kinematics: forward and inverse kinematics, DH parameters, joint spaces
  • Dynamics and control: PID control, model predictive control (MPC), impedance control
  • Imitation learning: behavioural cloning, DAgger, learning from demonstrations
  • Sim-to-real transfer: domain randomisation, system identification, reality gap
  • Reward shaping and curriculum learning for robotics
  • Manipulation: grasping (analytical, data-driven), dexterous manipulation, contact-rich tasks
  • Locomotion: legged robots, quadrupeds, humanoid balance, CPG-based control
  • Safety: safe exploration, constrained RL, risk-aware planning