摘要
针对移动机器人的未知环境下安全路径规划,论文采用了一种局部连接Hopfield神经网络(ANN)规划器。对任意形状环境,ANN中兼顾处理了“过近”和“过远”来形成安全路径,而无需学习过程。为在单处理器上进行有效的在线路径规划,提出用基于距离变换的串行模拟,加速了数值势场的传播。仿真表明该方法具有较高的实时性和环境适应性。
For the safe path planning of a mobile robot in unknown environments,the paper proposes a local linked Hopfield artificial neural network(ANN)planner.For environments of arbitrary shape,without learning process,the ANN plans a safe path with consideration of both″too close″and″too far″.For the effective application on sequential proces-sor to plan a path on-line,the simulation based on constrained distance transformation is proposed to accelerate the propagation of the numerical potential field of the ANN.Simulations demonstrats the method has high real-time ability and adaptability to environments.
出处
《计算机工程与应用》
CSCD
北大核心
2004年第8期86-89,共4页
Computer Engineering and Applications
关键词
移动机器人
安全路径规划
神经网络
约束距离变换
Mobile robot,Safe path planning,Neural networks,Constrained distance transformation
