Sébastien Kleff — accepted AI-conference papers

2025

Infinite-Horizon Value Function Approximation for Model Predictive Control

Armand Jordana, Sébastien Kleff, Arthur Haffemayer, Joaquim Ortiz de Haro, Justin Carpentier, Nicolas Mansard +1

RA-L 2025

Model Predictive Control has emerged as a popular tool for robots to generate complex motions. However, the real-time requirement has limited the use of hard constraints and large preview horizons, which are necessary to ensure safety and stability. In practice, practitioners have to carefully desig

Cited by 6 Source ↗Scholar ↗

2024

Force Feedback Model-Predictive Control via Online Estimation

Armand Jordana, Sébastien Kleff, Justin Carpentier, Nicolas Mansard, Ludovic Righetti

ICRA 2024poster

Nonlinear model-predictive control has recently shown its practicability in robotics. However it remains limited in contact interaction tasks due to its inability to leverage sensed efforts. In this work, we propose a novel model-predictive control approach that incorporates direct feedback from for…

Cited by 3 Source ↗Scholar ↗

2022

Introducing Force Feedback in Model Predictive Control

Sébastien Kleff, Ewen Dantec, Guilhem Saurel, Nicolas Mansard, Ludovic Righetti

IROS 2022poster

In the literature about model predictive control (MPC), contact forces are planned rather than controlled. In this paper, we propose a novel paradigm to incorporate effort measurements into a predictive controller, hence allowing to control them by direct measurement feedback. We first demonstrate w…

Cited by 14 Source ↗Scholar ↗

2022

Value learning from trajectory optimization and Sobolev descent: A step toward reinforcement learning with superlinear convergence properties

Amit Parag, Sébastien Kleff, Léo Saci, Nicolas Mansard, Olivier Stasse

ICRA 2022poster

The recent successes in deep reinforcement learning largely rely on the capabilities of generating masses of data, which in turn implies the use of a simulator. In particular, current progress in multi body dynamic simulators are under-pinning the implementation of reinforcement learning for end-to-…

Cited by 15 Source ↗Scholar ↗