Synthesis of Control Barrier Functions Using a Supervised Machine Learning Approach

Control barrier functions are mathematical constructs used to guarantee safety for robotic systems. When integrated as constraints in a quadratic programming optimization problem, instantaneous control synthesis with real-time performance demands can be achieved for robotics applications. Prevailing use has assumed full knowledge of the safety barrier functions, however there are cases where the safe regions must be estimated online from sensor measurements. In these cases, the corresponding barrier function must be synthesized online. This paper describes a learning framework for estimating control barrier functions from sensor data. Doing so affords system operation in unknown state space regions without compromising safety. Here, a support vector machine classifier provides the barrier function specification as determined by sets of safe and unsafe states obtained from sensor measurements. Theoretical safety guarantees are provided. Experimental ROS-based simulation results for an omnidirectional robot equipped with LiDAR demonstrate safe operation.

[1]  Geoffrey J. Gordon,et al.  A Reduction of Imitation Learning and Structured Prediction to No-Regret Online Learning , 2010, AISTATS.

[2]  Li Wang,et al.  Permissive Barrier Certificates for Safe Stabilization Using Sum-of-squares , 2018, 2018 Annual American Control Conference (ACC).

[3]  Siddharth Mayya,et al.  An Optimal Task Allocation Strategy for Heterogeneous Multi-Robot Systems , 2019, 2019 18th European Control Conference (ECC).

[4]  John C. Knight,et al.  Safety critical systems: challenges and directions , 2002, Proceedings of the 24th International Conference on Software Engineering. ICSE 2002.

[5]  Sherali Zeadally,et al.  Autonomous Cars: Research Results, Issues, and Future Challenges , 2019, IEEE Communications Surveys & Tutorials.

[6]  Sungwan Kim,et al.  A study on a robot arm driven by three-dimensional trajectories predicted from non-invasive neural signals , 2015, BioMedical Engineering OnLine.

[7]  Magnus Egerstedt,et al.  Control of Multi-Agent Systems with Finite Time Control Barrier Certificates and Temporal Logic , 2018, 2018 IEEE Conference on Decision and Control (CDC).

[8]  Aaron D. Ames,et al.  An Online Approach to Active Set Invariance , 2018, 2018 IEEE Conference on Decision and Control (CDC).

[9]  Nello Cristianini,et al.  Controlling the Sensitivity of Support Vector Machines , 1999 .

[10]  Michael C. Yip,et al.  Autonomous Navigation in Unknown Environments using Sparse Kernel-based Occupancy Mapping , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).

[11]  Fabio Tozeto Ramos,et al.  Hilbert maps: Scalable continuous occupancy mapping with stochastic gradient descent , 2015, Robotics: Science and Systems.

[12]  Barbara Hammer,et al.  A Note on the Universal Approximation Capability of Support Vector Machines , 2003, Neural Processing Letters.

[13]  Ettore Merlo,et al.  Engineering Safety in Swarm Robotics , 2018, 2018 IEEE/ACM 1st International Workshop on Robotics Software Engineering (RoSE).

[14]  Aaron D. Ames,et al.  Characterizing Safety: Minimal Barrier Functions from Scalar Comparison Systems , 2019, ArXiv.

[15]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[16]  Aaron D. Ames,et al.  Safety Barrier Certificates for Collisions-Free Multirobot Systems , 2017, IEEE Transactions on Robotics.

[17]  Paulo Tabuada,et al.  Control Barrier Functions: Theory and Applications , 2019, 2019 18th European Control Conference (ECC).

[18]  Andreas Krause,et al.  The Lyapunov Neural Network: Adaptive Stability Certification for Safe Learning of Dynamical Systems , 2018, CoRL.

[19]  Lionel Ott,et al.  Fast Stochastic Functional Path Planning in Occupancy Maps , 2019, 2019 International Conference on Robotics and Automation (ICRA).

[20]  Gábor Orosz,et al.  End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Tasks , 2019, AAAI.