Hierarchical Control
A robot control architecture that separates perception, decision-making and execution into different levels. Upper levels handle task planning while lower levels handle stable execution, reducing learning difficulty and improving reusability.
Timeline (last 90 days)
Hierarchical control typically has a high-level policy generate abstract subgoals or skill sequences, and a low-level policy convert the subgoals into concrete joint/end-effector actions.
The approach of combining hierarchical control with end-to-end learning is to keep an interpretable subgoal structure at the high level and use end-to-end models at the low level to improve responsiveness, thereby balancing long-horizon planning with real-time correction.
All Facts (2)
Hierarchical control typically has a high-level policy generate abstract subgoals or skill sequences, and a low-level policy convert the subgoals into concrete joint/end-effector actions.
50%UnverifiedThe approach of combining hierarchical control with end-to-end learning is to keep an interpretable subgoal structure at the high level and use end-to-end models at the low level to improve responsiveness, thereby balancing long-horizon planning with real-time correction.
50%