为满足水上无人测量平台低功耗、低成本的设计和应用需求，基于盲人寻路原理提出依靠位置信息与碰撞判断实现脱困路径规划的改进人工势场脱困算法。该算法解决了人工势场法自身目标不可达与局部最小值问题，并且提出了主动碰撞机制，该机制仿照盲人寻路原理，依靠不断碰撞判断障碍物位置并做出下一步运动方向调整。算法中使用椭圆法确定主动碰撞目标，使脱困路径简单有效；并且加入了地形还原机制，将已探索的障碍物连接起来，还原被困环境地形，防止脱困过程中探索非必要区域，提升脱困效率。本文算法与初步改进的人工势场算法仿真对比结果表明，本文算法在脱困路径长度与简洁性方面均体现出优越性。

An enhanced artificial potential field escape algorithm inspired by blind navigation principles is developed to improve energy efficiency and cost-effectiveness in water surface unmanned measuring platforms. Positional data and collision detection were used to address the inaccessibility of goals and local minimum issues in traditional methods. An active collision mechanism was introduced, allowing directional adjustments based on obstacle detection, similar to the blind's navigation strategy. Targets for active collisions were identified using an elliptical method to simplify escape routes. A terrain reconstruction mechanism was included to connect explored obstacles, restoring the topography of the trapped environment and avoiding unnecessary exploration. The algorithm was tested against the improved artificial potential field algorithm. The results show that the proposed approach achieves shorter and simpler escape paths. It is concluded that the algorithm improves escape efficiency and navigation performance.