In most multi-agent robotic systems, the absence of centralized or agent-to-agent communication poses challenges in maintaining system stability and ensuring a safe inter-agent distance. The core of this research lies in the development of an adaptive estimation-based control design that integrates two essential components: an estimator that accurately predicts the states of preceding vehicles or agents and a safety-critical controller. This controller, designed using Control Lyapunov Functions (CLFs) and Control Barrier Functions (CBFs), ensures operational safety within the autonomous system by compensating for potential estimation errors. The proposed estimator-controller framework is implemented in scenarios with and without agent-to-agent communication and time-varying leader velocity, demonstrating successful performance in maintaining safety and stability. Additionally, physics engine-based simulations reinforce both the practical viability of the proposed control framework in real-world cruise control situations and the controller’s adeptness at maintaining safety amidst realistic operating conditions. Moreover, various leader maneuvers are investigated for leader-follower formation control to test the controller’s robustness in maintaining formation while prioritizing safety for each agent.

Creator

• Bohara, Vishrut

Contributors

• Leahy, Kevin
• Pinciroli, Carlo
• Farzan, Siavash

Degree

• MS

Unit

• Robotics Engineering

Publisher

• Worcester Polytechnic Institute

Identifier

• etd-115361

Keyword

• Control Lyapunov Function
• Formation Control
• Control Barrier Function
• Multi-Agent Robotic Systems
• Safety-Critical Controller Design
• Adaptive Cruise Control for Autonomous Vehicles

Advisor

• Farzan, Siavash

Committee

• Leahy, Kevin
• Pinciroli, Carlo

Defense date

• 2023-12-06

Year

• 2023

Date created

• 2023-12-18

Resource type

• Thesis

Source

• etd-115361

Rights statement

• In Copyright

Last modified

• 2024-01-25