Precise robot control from imprecise but fast computations
Robot workers and autonomous vehicles rely on sensors to perceive their environment and software to process this information and decide what to do. In real-world situations, they need to compute commands as fast as possible. However, optimisation solvers in robots are designed to converge to a solution with high precision, which takes time. In this context, the Marie Skłodowska-Curie Actions (MSCA) project LP-NORM will investigate how low precision can be sufficient, and how this can be used to develop faster solvers. It will study what approximations are acceptable for the problem formulation and the solution. The project will focus on model predictive control and instantaneous linearised control, applied to a wide variety of systems, from buses to humanoid robots.
Fields of science
- HORIZON.1.2 - Marie Skłodowska-Curie Actions (MSCA) Main Programme
Call for proposalSee other projects for this call
Funding SchemeHORIZON-AG-UN - HORIZON Unit Grant
78153 Le Chesnay Cedex
See on map