论文部分内容阅读
针对在单一学习机制中,移动机器人自主导航一般只适用于静态场景,适应性差的问题,提出一种动态场景自适应导航方法.该方法通过激光测距仪(LRF)获取周围环境的距离信息,在基于增量判别回归(IHDR)算法的单一学习机制导航的基础上,提出了最远距离优先机制的局部避障环节.该导航方法克服了传统导航方法对环境模型的过度依赖,并且本文提出的基于最远距离优先机制的局部避障算法,解决了基于单一学习机制的导航方法对动态场景适应能力不足的问题.本文将动态场景自适应导航方法应用到了MT-R机器人中,与基于单一学习机制的导航方法进行了对比实验,并且运用提出的局部避障算法,对实验中的激光数据进行了算法性能分析.实验结果证实了该方法的可行性,并显示了该方法在动态场景下的良好表现.
Aiming at the problem that the adaptive navigation of mobile robots is only suitable for static scenes and poor adaptability in autonomous learning, a dynamic scene adaptive navigation method is proposed in this paper.In this method, the distance information of surrounding environment is obtained by laser range finder (LRF) Based on the single learning mechanism navigation based on incremental discriminant regression (IHDR) algorithm, this paper proposes a local obstacle avoidance mechanism based on the longest distance priority mechanism, which overcomes the over-reliance on the environment model by traditional navigation methods and proposes The local obstacle avoidance algorithm based on the furthest distance priority mechanism solves the problem that the navigation method based on the single learning mechanism has insufficient adaptability to the dynamic scene.This paper applies the dynamic scene adaptive navigation method to the MT-R robot, Learning mechanism of the navigation method of a comparative experiment, and the use of the proposed local obstacle avoidance algorithm for experimental laser data algorithm performance analysis.Experimental results confirm the feasibility of the method, and shows the method in dynamic scenarios Good performance.