基于模糊神经网络算法的机器人路径规划研究

Study on the Robot Path Planning Based on Fuzzy Neural Network Algorithm

路径规划是移动机器人研究的关键技术之一。在研究模糊理论和神经网络的基础上,提出了一种新的算法,即模糊神经网络。模糊神经网络既可以像神经网络那样并行处理、自行学习,又可以像模糊理论处理模糊信息、完成模糊推理功能。采用模糊神经网络来对移动机器人的路径进行规划,充分发挥模糊理论和神经网络的各自优势,从而获得从起始点到目标点的最优路径。在环境信息完全未知且静态的情况下进行了仿真实验,结果表明:该算法效率高、收敛速度快,有效提高了移动机器人的智能化水平。

Path planning is one of the key technologies for mobile robot researches. Based on the study of fuzzy theoryand neural network, a new algorithm namely fuzzy neural network is proposed. Fuzzy neural network-FNN can not only per-form parallel processing and self learning, as what NNs can do, but also process fuzzy information and perform fuzzy reason-ing, as what fuzzy theory can do. The fuzzy neural network-FNN is used for designing routes of mobile robots, and respectiveadvantages of fuzzy theory and NNs are fully utilized, then the optimum route from start point to object point is obtained. Inconditions where environmental information is unknown and static, simulation experiment is conducted and the results indi-cate that this algorithm is efficient and has fast convergence rate, which helps to improve the intelligence level of mobile ro-bots in an efficient way.