Using Model-Based Optimization and Motion Learning
The research proposed in this chair aims to generate complex motions for robots in real time by:
- relying on massive off-line caching of pre-computed optimal motions that are
- recovered and adapted online to new situations with real-time tractable model predictive control and where
- all available sensor modalities are exploited for feedback control going beyond the mere state of the robot for more robust behaviors.
The goal is to develop a unified yet tractable approach to motion generation for complex robots with arms and legs.
Programme : IA collaborative
Thèmes : Neuroscience et IA, Robotique et IA
Porteur :
Nicolas Mansard, DR CNRS, Laas
Contact
Nicolas.Mansard@laas.fr