首页> 外文会议>Evolutionary Computation, 2000. Proceedings of the 2000 Congress on >An extended virtual force field based behavioral fusion with neuralnetworks and evolutionary programming for mobile robot navigation
【24h】

An extended virtual force field based behavioral fusion with neuralnetworks and evolutionary programming for mobile robot navigation

机译:基于扩展虚拟力场的神经行为融合网络和用于移动机器人导航的进化编程

获取原文

摘要

A local navigation algorithm for mobile robots is proposed, basedon the new extended virtual force field (EVFF) concept, neuralnetwork-based fusion for the three primitive behaviors generated by theEVFF, and the evolutionary programming-based optimization of the neuralnetwork weights. Furthermore, a multi-network version of the aboveneurally-combined EVFF has been proposed that lends itself not only toan efficient architecture but also to a greatly enhanced generalizationcapability. These techniques have been verified through both simulationand real experiments under a collection of complex environments
机译:在此基础上,提出了一种用于移动机器人的局部导航算法。 关于新的扩展虚拟力场(EVFF)概念,神经 基于网络的融合,用于由网络生成的三种原始行为 EVFF,以及基于进化编程的神经网络优化 网络权重。此外,上述的多网络版本 已经提出了神经组合的EVFF,其不仅适用于 一个有效的体系结构,但也大大增强了概括性 能力。这些技术已通过两个模拟进行了验证 并在一系列复杂环境下进行实际实验

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号