Dynamic event-triggered cooperative tracking algorithm for high-speed vehicles

Abstract To address the challenges of single-sensor visual limitations and network congestion in tracking high-speed vehicles, a dynamic event-triggered cooperative cubature Kalman filtering algorithm is proposed. Initially, a dynamic event-triggered mechanism is developed based on measurement state information to reduce bandwidth consumption and enhance system responsiveness. Subsequently, within the framework of the cubature Kalman Filter, the covariance matrix and filtering gain matrix for each sensor’s distributed filter are derived using minimum mean square error approximation techniques. Finally, a cooperative filtering algorithm is designed using a weighted average consensus method and applied to tracking of hypersonic glide vehicles. Simulation results demonstrate that the proposed algorithm achieves higher accuracy compared to existing methods.