Cooperative Safe Trajectory Planning for Quadrotor Swarms

Area of Science:

• Robotics
• Control Systems
• Artificial Intelligence

Background:

• Cooperative trajectory planning for quadrotor swarms is crucial for complex missions.
• Existing methods often face challenges with scalability, real-time performance, and safety guarantees.
• Distributed control offers a promising approach to overcome limitations of centralized systems.

Purpose of the Study:

• To propose a novel distributed algorithm for cooperative trajectory planning in quadrotor swarms.
• To enhance safety and efficiency in large-scale quadrotor formations.
• To reduce communication overhead while maintaining robust performance.

Main Methods:

• Developed a distributed algorithm combining Model Predictive Control and Alternating Direction Method of Multipliers (DMPC-ADMM).
• Utilized differential flatness for nonlinear quadrotor dynamics and relaxed discrete-time control barrier functions (DCBF) for safety-feasibility balance.
• Implemented a peer-to-peer communication strategy with an event-triggered mechanism for reduced overhead.

Main Results:

• Generated real-time, safe, and smooth trajectories for quadrotor swarms.
• Achieved consensus on future trajectories among quadrotors within communication range, enhancing safety.
• Demonstrated superior performance compared to centralized and other distributed strategies in simulations.

Conclusions:

• The DMPC-ADMM algorithm is effective for cooperative trajectory planning in quadrotor swarms.
• The proposed method offers a scalable and efficient solution for real-time, safe, and smooth multi-quadrotor operations.
• The event-triggered mechanism successfully reduces communication load without compromising performance.