BarrierNet: Differentiable Control Barrier Functions for Learning of Safe Robot Control

Many safety-critical applications of neural networks, such as robotic control, require safety guarantees. This article introduces a method for ensuring the safety of learned models for control using differentiable control barrier functions (dCBFs). dCBFs are end-to-end trainable and guarantee safety...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on robotics 2023-06, Vol.39 (3), p.2289-2307
Main Authors: Xiao, Wei, Wang, Tsun-Hsuan, Hasani, Ramin, Chahine, Makram, Amini, Alexander, Li, Xiao, Rus, Daniela
Format: Article
Language:English
Subjects:
Autonomous vehicles
Changing environments
Control barrier function (CBF)
Controllers
Neural networks
Quadratic programming
Robot control
robot learning
Robot sensing systems
Robots
Safety
Safety critical
safety guarantees
Uncertainty
Vehicle dynamics
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Many safety-critical applications of neural networks, such as robotic control, require safety guarantees. This article introduces a method for ensuring the safety of learned models for control using differentiable control barrier functions (dCBFs). dCBFs are end-to-end trainable and guarantee safety. They improve over classical control barrier functions (CBFs), which are usually overly conservative. Our dCBF solution relaxes the CBF definitions by: 1) using environmental dependencies; 2) embedding them into differentiable quadratic programs. These novel safety layers are called a BarrierNet. They can be used in conjunction with any neural network-based controller. They are trained by gradient descent. With BarrierNet, the safety constraints of a neural controller become adaptable to changing environments. We evaluate BarrierNet on the following several problems: 1) robot traffic merging; 2) robot navigation in 2-D and 3-D spaces; 3) end-to-end vision-based autonomous driving in a sim-to-real environment and in physical experiments; 4) demonstrate their effectiveness compared to state-of-the-art approaches.
ISSN:1552-3098
1941-0468
DOI:10.1109/TRO.2023.3249564