A Markov Jump System-Based Mathematical Modeling Method for Collaborative Control of Unmanned Vehicles

The rapid development of smart vehicles has made collaborative control of unmanned vehicles a highly focused research field. Owing to the time-varying traffic context, collaborative control of unmanned vehicles always faces the challenges of complex and uncertain scenario characteristics. It is supposed to develop a robust mathematical modeling scheme that integrates these factors, for the above technical demand. Firstly, the Markov jump system is introduced as the core of mathematical modeling to extract information from vehicle space. Then, the collaborative control of driving conflicts and local conflicts on signalless road sections is fused and calculated. By capturing the transition probability of the system state, the behavior of unmanned vehicles was predicted, Based on the prediction results, corresponding control strategies are adopted. Finally, lane changing, following, obstacle warning, and automatic parking case analysis can be completed. To verify the effectiveness of the proposed method, we conducted a series of simulation experiments and compared them with traditional methods. Some typical instance cases including lane changing, vehicle following, obstacle warning, and automatic parking, are completed under collaborative control. The experimental results show that the proposed mathematical modeling method can significantly improve the prediction accuracy and effectiveness of delayed driving events, providing more reliable prediction capabilities for collaborative control of unmanned vehicle.