基于多传感器信息融合的AGV避障算法

基于多传感器信息融合的AGV避障算法

基于多传感器信息融合的AGV避障算法


摘要:介绍了多传感器信息融合技术的基本原理,研究了它在小车自主避障上的应用,并提出了在基于视觉定位的自动导引车(AGV)基础上的一种避障系统设计方案。该系统将超声波传感器与红外传感器组合使用,利用多传感器信息融合技术,采取改进的 K/N 融合方法以及基于先验知识的 AND 融合方法,对传感器数据进行分析决策并以此实现避障算法。实验利用上述方案进行避障测试,得到轨迹曲线。结果表明:该方案有效地改善了单一传感器性能较差以及消除噪声干扰等问题,避障时的安全距离不少于50 cm,可及早预警并采取相应避障措施,最终实现了较好的避障功能。
Abstract:Basic principle of multi-sensor information fusion technology and its application in vehicle obstacle avoidance were introduced,and a design of obstacle-avoiding system on vision-based positio-ning automatic guided vehicle (AGV)was put forward.With a combination of the ultrasonic sensor and the infrared sensor,this system on the base of multi-sensor information fusion technology applied the improved K/N fusion as well as the AND fusion with prior knowledge to collect relative data, which could be analyzed by decision-making rules in order to achieve the obstacle-avoiding algorithm. The experiment using the above-mentioned design for testing obtained the trajectory curve.The result shows that the designed application can effectively improve the poor performance of the single sensor and eliminate the risk of noise interference.The safe distance is not less than 50 cm and can take ap-propriate avoiding measures with early warning.As a consequence,this application of obstacle avoid-ance function in multi–senor data fusion technology achieves a desired result.
doi:10.13245/j.hust.15S1054
作者:李沛 李新德
Author:Li Pei Li Xinde
作者单位: 东南大学自动化学院,江苏 南京,210096
刊 名:华中科技大学学报(自然科学版) ISTICEIPKU
Journal:Journal of Huazhong University of Science and Technology (Nature Science Edition)
年,卷(期):2015, (z1)
分类号:TP242.6
关键词:自动导引车知识脉络超声波传感器知识脉络红外传感器知识脉络信息融合知识脉络避障知识脉络
Keywords:automatic guided vehiclesultrasonic sensorinfrared sensorformation fusionobstacle avoidance
在线出版日期:2015年11月2日
基金项目:国家自然科学基金资助项目(60804063,61175091,61573097);江苏省“青蓝工程”及“六大高峰人才”资助计划.

下载地址:
链接:https://pan.baidu.com/s/1c2AMf0c 密码:3z54

© 版权声明
THE END
喜欢就支持一下吧
点赞0 分享
评论 抢沙发

请登录后发表评论

    暂无评论内容