Autonomous-Mobile Robot Navigation under Dynamic Path Environments

动态路径环境下的自主移动机器人导航

• 批准号: 12650437
• 负责人: TSUCHIYA Takeshi
• 金额: $ 2.11万
• 依托单位: HOKKAIDO UNIVERSITY
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 2000
• 资助国家: 日本
• 起止时间: 2000 至 2001
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-12650437/
• 关键词: Mobile Robot; Navigation; Path Planning; Nonliaear Filter; Ultrasonic sensor; Fuzzy Reasoning; 自律ロボット; 環境認識

PORPOSE OF THE PROJECT : [REALIZATION OF MOBILE ROBOT WHICH CAN RECOGNIZE THEIR ENVIRONMENTS BY THEMSELVES AND PLAN THEIR ACTIONS ON THE BASIS OF THE OBTAINED KNOWLEDGE ]*Development of Navigation Algorithm for Autonomous Mobile Robot and Robot Experiments :Rand-Marks at the intersections on the path are assumed in our mobile robot experiments. At the first stage of the experiments; robots run about the path randomly in order to obtain the knowledge about the environments. Robots plan their actions on the basis of the obtained knowledge about the environments. Robots are able to avoid the sudden change of the environments if it is not so big change using their reflex-like movements :*Propose of Nonlinear Filter for Robot Sensor :Ultrasonic sensors are widely utilized for robots because of the non-expensive sensors but signals obtained by the sensors contain many undesirable noises, especially impulsive noise. New nonlinear filter based on sliding mode system is developed and the effectiveness of the filter is appealed for the existed many filter including linear and nonlinear filters.*Propose of Distance-type Fuzzy Reasoning Method :New Fuzzy reasoning based on the distance between each fuzzy sets, which overcome drawbacks arisen from the usual fuzzy reasoning. A new rubber string-type associated memory algorithm and a new learning algorithm are developed on the basis of the distance-type fuzzy reasoning. These approaches may useful for the autonomous mobile robot project.

项目目的：[实现移动机器人，它可以通过自己识别他们的环境，并在获得的知识的基础上进行他们的行动]* 自主移动的机器人和机器人实验的导航算法的开发：在我们的移动的机器人实验中，在路径上的交叉点处假设随机标记。在实验的第一阶段，机器人随机地在路径上运行，以获得有关环境的知识。机器人根据所获得的环境知识来规划它们的行动。机器人能够避免环境的突然变化，如果它不是那么大的变化，使用他们的反射式运动：* 机器人传感器的非线性滤波器的建议：超声波传感器被广泛用于机器人，因为不昂贵的传感器，但传感器获得的信号包含许多不希望的噪声，特别是脉冲噪声。针对已有的线性和非线性滤波器，提出了一种新的基于滑模系统的非线性滤波器，并对该滤波器的有效性进行了分析。距离型模糊推理方法的提出：基于各模糊集之间的距离进行模糊推理，克服了通常模糊推理方法的缺点。在距离型模糊推理的基础上，提出了一种新的橡胶串型联想记忆算法和一种新的学习算法。这些方法对自主式移动的机器人项目有一定的参考价值。

项目成果

T. Emaru: "Avoid Crosstalk of Sonar by Using Sliding Mode System"Proceedings of IECON (Nagoya). 456-466 (2000)

T. Emaru：“使用滑模系统避免声纳串扰”IECON 会议记录（名古屋）。

Shoyu Wang: "A Learning Algorithm for Discrete-type Fuzzy Reasoning Method"Biomedical Soft Computing And Human Sciences. 6.1. 61-68 (2000)

王守宇：《一种离散型模糊推理方法的学习算法》生物医学软计算与人文科学。

T. Tsuchiya: "New Edition Modern Control Engineering"Sangyo Tosho Inc.. (2000)

T. Tsuchiya：“新版现代控制工程”Sangyo Tosho Inc.（2000）

江上 正: "最適予見制御と一般化予測制御"計測自動制御学会誌. 39・5. 337-342 (2000)

江上正：“最优预测控制和广义预测控制”仪器与控制工程师学会杂志39・5（2000）。

王 碩玉: "距離型ファジイ推論法"バイオメディカル・ファジィシステム学会誌. 1.1. 61-78 (1999)

王硕宇：“距离型模糊推理方法”生物医学模糊系统学会杂志1.1（1999）。