Implicit Subgoal Planning with Variational Autoencoders for Long-Horizon Sparse Reward Robotic Tasks.
Fangyuan WangAnqing DuanPeng ZhouShengzeng HuoGuodong GuoChenguang YangDavid Navarro-AlarconPublished in: CoRR (2023)
Keyphrases
- robotic tasks
- motion planning
- planning under uncertainty
- robotic systems
- goal oriented
- dynamic environments
- belief space
- mobile robot
- reinforcement learning
- degrees of freedom
- denoising
- ai planning
- vision system
- humanoid robot
- high dimensional
- real time
- planning problems
- dynamical systems
- integer programming
- long run
- video sequences