HM-Science
Human-machine shared control for intelligent vehicle safety and energy
The main objective of the HM-Science (Human-Machine Shared Control for Intelligent Safety and Energy of Smart Vehicles) project is to develop new control approaches and architectures for "smart" vehicles, in order to design human-centered shared control systems.
A lot of work in the field of autonomous driving has shown that trying to define behaviors suitable for all users (including People with Reduced Mobility - PRM) and adapted to all situations that may be encountered, is very complex, if not impossible. The aim of HM-Science is to share and, to do so, build H-M cooperation; in particular, to give machines learning capabilities, to enable them to learn and adapt to users and situations.
The solution developed is at the crossroads of two important fields - control and AI - and the key challenge will be to be able to combine them in a framework that preserves safety and real-time performance.
This project fits perfectly into the important questioning of these disciplines as pointed out by (Recht 2019 (*)): "A final important problem, which could be the most daunting of all, is how machines should learn when humans are in the loop. What can humans interacting with robots do, and how can human actions be modeled?"
(*) Recht B. (2019). A tour of reinforcement learning: The view from continuous control. Annual Review of Control, Robotics, and Autonomous Systems 2, 253-279
| Department(s) | Partner(s) | Overall amount |
|---|---|---|
| Automatique |
278 k€
|
|
| Main support | Rayout | Date(s) |
| ANR | International |
2022 - 2024
|