Abstract: To improve the solving accuracy and local extremum escape ability of chimpanzee optimization algorithm (ChOA), this paper proposes a street lamp shadow imaging and running away from home strategy for improved chimpanzee optimization algorithm (SSR-ChOA). Firstly, the population is initialized using SPM chaotic sequences to increase the uniformity of the initial population distribution. Secondly, this paper designs a new optical improvement method based on the physical phenomenon of human shadow changes under street lights at night: street lamp shadow imaging strategy. This strategy is used to optimize the problem of low development accuracy of ChOA algorithm. This paper designs a global optimization strategy called running away from home, which enables ordinary chimpanzee individuals to have stronger proactive exploration abilities. This strategy can help individual chimpanzees to jump out of local extrema caused by leader wrong judgement, avoiding stagnation and premature convergence in population search. This paper tests 25 benchmark test functions and CEC2014 test functions. The ChOA algorithm, 4 different types of improved ChOA algorithms, and particle swarm optimization algorithm are compared. This paper analyzes the effectiveness of the improvement strategy. Finally, the application scenarios of towering electric towers and signal towers in the flight path of aerial drones are studied. This paper verifies the effectiveness of SSR-ChOA. Experimental results show that SSR-ChOA has significant differences compared with ChOA and 4 improved ChOA, and SSR-ChOA has significant advantages in optimization accuracy and stability. In terms of 3D path planning for UAV, the average total cost of SSR-ChOA is 3.06% lower than that of ChOA.

Key words: chimpanzee optimization algorithm (ChOA), SPM chaotic sequence, street lamp shadow imaging strategy, running away from home strategy, 3D path planning for UAV

摘要： 为提高黑猩猩优化算法（ChOA）的求解精度和局部极值逃逸能力，提出一种路灯人影和离家出走改进的黑猩猩优化算法（SSR-ChOA）。首先，采用SPM混沌序列初始化种群，增加初始种群分布均匀性。其次，由夜间路灯下人影变化的物理现象设计一种新的光学类改进方式——路灯人影，用于优化原有ChOA算法开发精度不高问题。同时设计一种名为离家出走的全局优化策略，使普通黑猩猩个体拥有更强的主动探索能力，避免因领导者判断错误陷入局部极值而导致整个种群搜索停滞、过早收敛。测试了25个基准测试函数和CEC2014测试函数，对比了ChOA算法、4种不同类型改进ChOA算法以及粒子群等算法，分析了改进策略有效性。最后，对航拍无人机飞行路径中存在高耸电塔、信号塔的应用情景进行了研究，验证了SSR-ChOA有效性。实验结果表明，SSR-ChOA与ChOA和4种改进ChOA对比不仅具有显著性差异，而且在寻优精度和稳定性上表现更佳。无人机3D路径规划上，SSR-ChOA平均总开销相比ChOA减少3.06%。

关键词: 黑猩猩优化算法（ChOA）, SPM混沌序列, 路灯人影策略, 离家出走策略, 无人机3D路径规划