TY - JOUR
T1 - Generalization of Safe Optimal Control Actions on Networked Multiagent Systems
AU - Song, Lin
AU - Wan, Neng
AU - Gahlawat, Aditya
AU - Tao, Chuyuan
AU - Hovakimyan, Naira
AU - Theodorou, Evangelos A.
N1 - Funding Information:
This work was supported in part by the Air Force Office of Scientific Research (AFSOR), in part by National Aeronautics and Space Administration (NASA), and in part by National Science Foundation's National Robotics Initiative (NRI) and Cyber-Physical Systems (CPS) under Grant 1830639, Grant 1932529, and Grant 1932288.
Publisher Copyright:
© 2014 IEEE.
PY - 2023/3/1
Y1 - 2023/3/1
N2 - In this article, we propose a unified framework to instantly generate a safe optimal control action for a new task from existing controllers on multiagent systems. The control action composition is achieved by taking a weighted mixture of the existing controllers according to the contribution of each component task. Instead of sophisticatedly tuning the cost parameters and other hyperparameters for safe and reliable behavior in the optimal control framework, the safety of each single-task solution is guaranteed using the control barrier functions (CBFs) for high relative degree stochastic systems, which constrains the system state within a known safe operation region where it originates from. Linearity of CBF constraints in control ensures the feasibility of safe control action composition. The discussed framework can immediately provide reliable solutions to new tasks by taking a weighted mixture of solved component-task actions and satisfying some CBF constraints, instead of performing an extensive sampling to compute a new controller. Our results are verified and demonstrated on both a single unmanned aerial vehicle (UAV) and two cooperative UAV teams in an environment with obstacles.
AB - In this article, we propose a unified framework to instantly generate a safe optimal control action for a new task from existing controllers on multiagent systems. The control action composition is achieved by taking a weighted mixture of the existing controllers according to the contribution of each component task. Instead of sophisticatedly tuning the cost parameters and other hyperparameters for safe and reliable behavior in the optimal control framework, the safety of each single-task solution is guaranteed using the control barrier functions (CBFs) for high relative degree stochastic systems, which constrains the system state within a known safe operation region where it originates from. Linearity of CBF constraints in control ensures the feasibility of safe control action composition. The discussed framework can immediately provide reliable solutions to new tasks by taking a weighted mixture of solved component-task actions and satisfying some CBF constraints, instead of performing an extensive sampling to compute a new controller. Our results are verified and demonstrated on both a single unmanned aerial vehicle (UAV) and two cooperative UAV teams in an environment with obstacles.
KW - Control barrier functions (CBFs)
KW - multiagent systems (MASs)
KW - safe control
KW - stochastic optimal control
UR - http://www.scopus.com/inward/record.url?scp=85137568998&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85137568998&partnerID=8YFLogxK
U2 - 10.1109/TCNS.2022.3203479
DO - 10.1109/TCNS.2022.3203479
M3 - Article
AN - SCOPUS:85137568998
SN - 2325-5870
VL - 10
SP - 491
EP - 502
JO - IEEE Transactions on Control of Network Systems
JF - IEEE Transactions on Control of Network Systems
IS - 1
ER -