Path planning is crucial for robotics, enabling robots to find collision-free routes from their current positions to target positions. The Artificial Potential Field (APF) approach utilizes attractive and repulsive fields to guide robots towards targets while avoiding obstacles. However, the conventional APF's repulsive potential equation can yield suboptimal results due to local minima. To address this, a novel method called Multi-Objective Evolutionary Artificial Potential Field (MOE-APF) is introduced. MOE-APF modifies the repulsive potential equation and employs the membrane computing and Genetic Algorithm (GA)to optimize a newset of APF parameters. The fitness function considers multiple objectives: path length, smoothness, success rate, and safety. A comparison witha recent method calledMembrane Evolutionary Artificial Potential Field (memEAPF) shows that MOE-APF significantly enhances path quality, optimization time, and success rate across various environments. MOE-APF's versatility allows it to tackle path planning challenges involving non-holonomic robots, multiple robots, industrial manipulators, and dynamic obstacles