EventSession Formset ID 287

Event: ccnw23 2023
Type: Work Group
Title: Balancing real or simulation cart-pole system with RL
is_program_session: True
Description:

Equipment: Cart-pole simulator and real cart-pole used by Tobi Delbruck for MPC tests

 

Possible directions:

  • Online Spiking RL (R-Eprop) for non-ideal cartpole stabilization solving in simulation after transfer learning of weights for ideal cartpole trained with DQN 
  • Deep Q-Learning for cartpole swing-up in simulation
  • Implementation of inference DQN on real cartpole
  • Possible simulation of cartpole with event-based tracked input instead of positions + velocities

 

Possible Challenges:

  • Use an event-based tracking algorithm to compute the shape coordinates over time
  • Define the RL policy (e.g., actions, state, reward) in simulation based on an event-camera output
  • Transfer the policy to the real cart-pole
  • Learn how to control the cart-pole
Speaker:
Moderators: Maxime Fabre, Luna Gava, Shreyan Banerjee,
Schedule ID start time end time location
347 May 02 2023, 22:00 May 02 2023, 23:00 Lobby
356 May 03 2023, 16:00 May 03 2023, 17:00 Disco
List all sessions Update Delete