森林火灾自主监测无人机的三维混沌路径规划与跟踪控制研究

Forest fire is a serious disaster that destroys the forest and threatens the ecological environment. Fire detection early could reduce or avoid the occurrence of harmful accidents. The unmanned aerial vehicle (UAV) is used easily for the independent monitoring of forest fires because of its rapid mobility, low cost and simple operation. However, there exist some problems such as poor path randomness, low coverage, high overlap rate and poor real-time path tracking. Therefore，based on the theoretical research of super chaotic system, this project intends to raise the stochastic full coverage three-dimensional path planning and tracking control of UAV for autonomously and effectively monitoring forest fires. Firstly, the super chaotic system with strong ergodicity and high randomness will be obtained by analyzing and controlling the memristor multi-scroll chaotic system. Based on the theories, the stochastic full coverage two-dimensional path planning algorithm and the random path algorithm of patrolling the key regions will be given, which could solve the problem of low coverage, high overlap rate and poor randomness. Then, the stochastic full coverage three-dimensional chaotic path planning algorithm will be proposed by combining with forest geographic information. Finally, the universal real-time tracking controller of the UAV with multiple constraints will be designed to improve the generality and real-time performance of path tracking, which will achieve autonomous random full coverage inspection and key regions monitor for the forest fire. This project not only deepens the research of three-dimensional chaotic path planning, but also improves the intelligence and early warning capability of forest fire prevention. Therefore, it has important theoretical innovation and broad application prospects.

森林火灾是破坏森林、危害生态环境的严重灾害，火情早发现可减少或避免危害性事故的发生。无人机因机动快速、成本低、操作简单等优点被用于森林火灾的自主监测，但存在路径随机性差、覆盖率低、重叠率高、路径跟踪实时性差等问题。因此，本项目拟基于超级混沌系统的理论研究，开展无人机自主高效监测森林火灾的随机全覆盖三维混沌路径规划与跟踪控制研究。首先，分析与控制构造的忆阻多涡卷混沌系统，获得强遍历性和高随机性的超级混沌理论；以此理论为基础，研究重点区域必巡的随机全覆盖二维路径规划算法和随机路径算法，解决覆盖率低、重叠率高、随机性差的问题；然后，结合森林地理信息提出三维混沌路径规划算法；最后，设计多约束条件下通用航迹实时跟踪控制器，提高路径跟踪的通用性和实时性，实现无人机自主随机全覆盖巡查和重点监控。本研究深化了三维路径规划的理论研究，提高了森林防火的智能化和早预警能力，具有重要的理论创新和广阔的应用前景。

森林火灾是破坏森林、危害生态环境的严重灾害，火情早发现可减少或避免危害性事故的发生。无人机因机动快速、成本低、操作简单等优点被用于森林火灾的自主监测，但存在路径随机性差、覆盖率低、重叠率高、路径跟踪实时性差等问题。因此，本项目基于超级混沌系统的理论研究，开展无人机自主高效监测森林火灾的随机全覆盖三维混沌路径规划与跟踪控制研究。首先，分析与控制构造的忆阻多涡卷混沌系统，获得了强遍历性和高随机性的超级混沌理论；以此理论为基础，研究了随机全覆盖路径规划算法，解决覆盖率低、重叠率高、随机性差的问题；然后，针对森林火灾不易发生季节，根据火灾发生等级，引入约束条件，规划了重点区域必巡查的随机路径，监测森林火灾效率更高；最后，设计了多约束条件下通用航迹实时跟踪控制器，提高路径跟踪的通用性、实时性和稳定性，实现无人机自主随机全覆盖巡查和重点监控。本研究深化了路径规划的理论研究，有效解决森林火灾的发生时间和地点是随机和不确定的问题，提高了森林防火的智能化和早预警能力，具有重要的理论创新和广阔的应用前景。