周 俊,陈 钦,梁 泉.基于强化学习的农业移动机器人视觉导航[J].农业机械学报,2014,45(2):53-58.
Zhou Jun,Chen Qin,Liang Quan.Vision Navigation of Agricultural Mobile Robot Based on Reinforcement Learning[J].Transactions of the Chinese Society for Agricultural Machinery,2014,45(2):53-58.
摘要点击次数: 4227
全文下载次数: 1892
基于强化学习的农业移动机器人视觉导航   [下载全文]
Vision Navigation of Agricultural Mobile Robot Based on Reinforcement Learning   [Download Pdf][in English]
投稿时间:2012-12-23  
DOI:10.6041/j.issn.1000-1298.2014.02.010
中文关键词:  农业机器人 强化学习 模糊逻辑 视觉导航
基金项目:国家自然科学基金资助项目(31071325)和江苏省自然科学基金资助项目(BK2010458)
作者单位
周 俊 南京农业大学 
陈 钦 南京农业大学 
梁 泉 南京农业大学 
中文摘要:以强化学习为基础,结合模糊逻辑理论研究了农业移动机器人通过自主学习获取导航控制策略的方法。首先使用机器视觉检测环境障碍并获取障碍物相对于移动机器人的方向和距离信息。然后应用强化学习设计了机器人自主获取导航控制策略方法,使机器人能够不断适应动态变化的导航环境。最后基于模糊逻辑离散化连续的障碍物方向和距离信息,构建了离散化的环境状态,并据此制定了自主导航学习Q值表。在自制的轮式移动机器人平台上开展了试验,结果表明机器人可以在实际导航环境中自动获取更优的导航策略,完成预期的导航任务。
Zhou Jun  Chen Qin  Liang Quan
Nanjing Agricultural University;Nanjing Agricultural University;Nanjing Agricultural University
Key Words:Agricultural robot Reinforcement learning Fuzzy logic Vision navigation
Abstract:The method that agricultural mobile robot acquire the navigation strategies through autonomous learning was development based on reinforcement learning and fuzzy logic. Firstly, the machine vision was applied to detect obstacles in the navigation environment, and the corresponding direction and distance between the robot and the obstacle was calculated. Then the algorithm of acquiring the more optimal navigation strategies was introduced with the reinforcement learning, so the capability of the mobile robot of adapting the dynamic navigation environment was improved. Finally, the continuous values of the direction and the distance between the obstacles and the mobile robot were discretized with the fuzzy logic rules, and the discrete navigation environment states were obtained, then the Q value table was designed for the reinforcement learning. The experiment was carried out with the wheeled mobile robot, and the experimental results showed that the mobile robot was able to automatically acquire more optimal navigation strategies in the actual environment, and fulfill the expected navigation tasks.

Transactions of the Chinese Society for Agriculture Machinery (CSAM), in charged of China Association for Science and Technology (CAST), sponsored by CSAM and Chinese Academy of Agricultural Mechanization Science(CAAMS), started publication in 1957. It is the earliest interdisciplinary journal in Chinese which combines agricultural and engineering. It always closely grasps the development direction of agriculture engineering disciplines and the published papers represent the highest academic level of agriculture engineering in China. Currently, nearly 8,000 papers have been already published. There are around 3,000 papers contributed to the journal each year, but only around 600 of them will be accepted. Transactions of CSAM focuses on a wide range of agricultural machinery, irrigation, electronics, robotics, agro-products engineering, biological energy, agricultural structures and environment and more. Subjects in Transactions of the CSAM have been embodied by many internationally well-known index systems, such as: EI Compendex, CA, CSA, etc.

   下载PDF阅读器