An Improved Crested Porcupine Algorithm for Path Planning of Underwater Robot Fish

This paper proposes an Improved Crested Porcupine Optimizer (ICPO) to enhance the efficiency and accuracy of path planning for underwater bionic fish—enabling more effective obstacle avoidance, foraging, and predator evasion. Specifically, ICPO integrates an elite reverse learning mechanism to enhance global search capability and reduce the risk of premature convergence. An improved initialization strategy accelerates the early exploration phase, while a dynamic radius contraction mechanism increases adaptability and solution accuracy during the convergence stage. Experimental results demonstrate that ICPO exhibits superior performance and faster convergence. ICPO consistently identifies optimal paths with fewer iterations, reduced path lengths, and improved computational efficiency, underscoring its suitability for underwater navigation. Its robust performance across varied environmental conditions further affirms its effectiveness and reliability as a path-planning algorithm for bionic fish.