Learning Robust Output Control Barrier Functions From Safe Expert Demonstrations
This paper addresses learning safe output feedback control laws from partial observations of expert demonstrations. We assume that a model of the system dynamics and a state estimator are available along with corresponding error bounds, e.g., estimated from data in practice. We first propose robust...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Journal of Control Systems |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10491341/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841554025786376192 |
|---|---|
| author | Lars Lindemann Alexander Robey Lejun Jiang Satyajeet Das Stephen Tu Nikolai Matni |
| author_facet | Lars Lindemann Alexander Robey Lejun Jiang Satyajeet Das Stephen Tu Nikolai Matni |
| author_sort | Lars Lindemann |
| collection | DOAJ |
| description | This paper addresses learning safe output feedback control laws from partial observations of expert demonstrations. We assume that a model of the system dynamics and a state estimator are available along with corresponding error bounds, e.g., estimated from data in practice. We first propose robust output control barrier functions (ROCBFs) as a means to guarantee safety, as defined through controlled forward invariance of a safe set. We then formulate an optimization problem to learn ROCBFs from expert demonstrations that exhibit safe system behavior, e.g., data collected from a human operator or an expert controller. When the parametrization of the ROCBF is linear, then we show that, under mild assumptions, the optimization problem is convex. Along with the optimization problem, we provide verifiable conditions in terms of the density of the data, smoothness of the system model and state estimator, and the size of the error bounds that guarantee validity of the obtained ROCBF. Towards obtaining a practical control algorithm, we propose an algorithmic implementation of our theoretical framework that accounts for assumptions made in our framework in practice. We validate our algorithm in the autonomous driving simulator CARLA and demonstrate how to learn safe control laws from simulated RGB camera images. |
| format | Article |
| id | doaj-art-3c2fe38e54d74bdf8cb9d1e737ce7f38 |
| institution | Kabale University |
| issn | 2694-085X |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of Control Systems |
| spelling | doaj-art-3c2fe38e54d74bdf8cb9d1e737ce7f382025-01-09T00:03:00ZengIEEEIEEE Open Journal of Control Systems2694-085X2024-01-01315817210.1109/OJCSYS.2024.338534810491341Learning Robust Output Control Barrier Functions From Safe Expert DemonstrationsLars Lindemann0https://orcid.org/0000-0003-3430-6625Alexander Robey1https://orcid.org/0009-0003-5693-2819Lejun Jiang2https://orcid.org/0000-0001-9051-2218Satyajeet Das3https://orcid.org/0009-0001-6075-2664Stephen Tu4https://orcid.org/0000-0001-5178-8326Nikolai Matni5https://orcid.org/0000-0003-4936-3921Thomas Lord Department of Computer Science, University of Southern California, Los Angeles, CA, USADepartment of Electrical and Systems Engineering, University of Pennsylvania, Philadelphia, PA, USANuro, Inc., Mountain View, CA, USADepartment of Electrical and Systems Engineering, University of Pennsylvania, Philadelphia, PA, USAMing Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, USADepartment of Electrical and Systems Engineering, University of Pennsylvania, Philadelphia, PA, USAThis paper addresses learning safe output feedback control laws from partial observations of expert demonstrations. We assume that a model of the system dynamics and a state estimator are available along with corresponding error bounds, e.g., estimated from data in practice. We first propose robust output control barrier functions (ROCBFs) as a means to guarantee safety, as defined through controlled forward invariance of a safe set. We then formulate an optimization problem to learn ROCBFs from expert demonstrations that exhibit safe system behavior, e.g., data collected from a human operator or an expert controller. When the parametrization of the ROCBF is linear, then we show that, under mild assumptions, the optimization problem is convex. Along with the optimization problem, we provide verifiable conditions in terms of the density of the data, smoothness of the system model and state estimator, and the size of the error bounds that guarantee validity of the obtained ROCBF. Towards obtaining a practical control algorithm, we propose an algorithmic implementation of our theoretical framework that accounts for assumptions made in our framework in practice. We validate our algorithm in the autonomous driving simulator CARLA and demonstrate how to learn safe control laws from simulated RGB camera images.https://ieeexplore.ieee.org/document/10491341/Control barrier functionsdata-driven robust controloutput feedback control |
| spellingShingle | Lars Lindemann Alexander Robey Lejun Jiang Satyajeet Das Stephen Tu Nikolai Matni Learning Robust Output Control Barrier Functions From Safe Expert Demonstrations IEEE Open Journal of Control Systems Control barrier functions data-driven robust control output feedback control |
| title | Learning Robust Output Control Barrier Functions From Safe Expert Demonstrations |
| title_full | Learning Robust Output Control Barrier Functions From Safe Expert Demonstrations |
| title_fullStr | Learning Robust Output Control Barrier Functions From Safe Expert Demonstrations |
| title_full_unstemmed | Learning Robust Output Control Barrier Functions From Safe Expert Demonstrations |
| title_short | Learning Robust Output Control Barrier Functions From Safe Expert Demonstrations |
| title_sort | learning robust output control barrier functions from safe expert demonstrations |
| topic | Control barrier functions data-driven robust control output feedback control |
| url | https://ieeexplore.ieee.org/document/10491341/ |
| work_keys_str_mv | AT larslindemann learningrobustoutputcontrolbarrierfunctionsfromsafeexpertdemonstrations AT alexanderrobey learningrobustoutputcontrolbarrierfunctionsfromsafeexpertdemonstrations AT lejunjiang learningrobustoutputcontrolbarrierfunctionsfromsafeexpertdemonstrations AT satyajeetdas learningrobustoutputcontrolbarrierfunctionsfromsafeexpertdemonstrations AT stephentu learningrobustoutputcontrolbarrierfunctionsfromsafeexpertdemonstrations AT nikolaimatni learningrobustoutputcontrolbarrierfunctionsfromsafeexpertdemonstrations |