Safe and Fast Tracking on a Robot Manipulator: Robust MPC and Neural Network Control

Fast feedback control and safety guarantees are essential in modern robotics. We present an approach that achieves both by combining novel robust model predictive control (MPC) with function approximation via (deep) neural networks (NNs). The result is a new approach for complex tasks with nonlinear...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE robotics and automation letters Ročník 5; číslo 2; s. 3049 - 3056
Hlavní autoři: Nubert, Julian, Kohler, Johannes, Berenz, Vincent, Allgower, Frank, Trimpe, Sebastian
Médium: Journal Article
Jazyk:angličtina
Vydáno: Piscataway IEEE 01.04.2020
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Algorithms
Constraints
Deep learning in robotics and automation
Dynamic stability
Electron tubes
Feedback control
Manipulators
motion control
Network control
Neural networks
optimization and optimal control
Predictive control
redundant robots
Robot arms
Robotics
Robust control
robust/adaptive control of robotic systems
Robustness
Run time (computers)
Safety
Statistical methods
Task analysis
Task complexity
Tracking
Uncertainty
ISSN:2377-3766, 2377-3766
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Fast feedback control and safety guarantees are essential in modern robotics. We present an approach that achieves both by combining novel robust model predictive control (MPC) with function approximation via (deep) neural networks (NNs). The result is a new approach for complex tasks with nonlinear, uncertain, and constrained dynamics as are common in robotics. Specifically, we leverage recent results in MPC research to propose a new robust setpoint tracking MPC algorithm, which achieves reliable and safe tracking of a dynamic setpoint while guaranteeing stability and constraint satisfaction. The presented robust MPC scheme constitutes a one-layer approach that unifies the often separated planning and control layers, by directly computing the control command based on a reference and possibly obstacle positions. As a separate contribution, we show how the computation time of the MPC can be drastically reduced by approximating the MPC law with a NN controller. The NN is trained and validated from offline samples of the MPC, yielding statistical guarantees, and used in lieu thereof at run time. Our experiments on a state-of-the-art robot manipulator are the first to show that both the proposed robust and approximate MPC schemes scale to real-world robotic systems.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2377-3766
2377-3766
DOI:10.1109/LRA.2020.2975727