Scholarlink

<p>In this paper, the problem of real-time trajectory planning for unmanned aerial vehicle under multi-constraints in dynamic and complex environments is addressed. In order to solve significant variations in the attitude angle of existing algorithms, the explicit yaw angle model and velocity field are established to ensure the smoothness and continuity of the trajectory. By introducing an adaptive cost function and convex model predictive optimization, a dynamic time-varying trajectory is generated within the sampling time interval, ensuring precise reaching of the target position. The algorithm’s primary advantage lies in its rapid convergence properties and computational efficiency, achieving an 18.65% reduction in processing time compared to other algorithms. The simulation results demonstrate that trajectory planning and collision avoidance are effectively achieved in an environment consisting of randomly assigned and dynamically changing obstacles. In addition, the smoothness index is improved by 3.95%, which facilitates subsequent tracking control.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Adaptive Convex Model Predictive Trajectory Planning Algorithm Based on Velocity Field

  • 期刊论文

  • 发表日期：

    2025年4月25日

查看全文
  • Keqing Guo,
  • Hui Wang,
  • Heting Wang,
  • Xin Dai

摘要

In this paper, the problem of real-time trajectory planning for unmanned aerial vehicle under multi-constraints in dynamic and complex environments is addressed. In order to solve significant variations in the attitude angle of existing algorithms, the explicit yaw angle model and velocity field are established to ensure the smoothness and continuity of the trajectory. By introducing an adaptive cost function and convex model predictive optimization, a dynamic time-varying trajectory is generated within the sampling time interval, ensuring precise reaching of the target position. The algorithm’s primary advantage lies in its rapid convergence properties and computational efficiency, achieving an 18.65% reduction in processing time compared to other algorithms. The simulation results demonstrate that trajectory planning and collision avoidance are effectively achieved in an environment consisting of randomly assigned and dynamically changing obstacles. In addition, the smoothness index is improved by 3.95%, which facilitates subsequent tracking control.