基于多传感器信息融合的迎宾机器人避障问题研究

发布时间：2018-03-30 10:10

本文选题：迎宾机器人　切入点：信息融合　出处：《山东科技大学》2017年硕士论文

【摘要】：随着机器人和人工智能技术的发展,智能化成为机器人发展的一个主题。迎宾机器人要想实现智能化发展,必须解决自主避障问题。因为迎宾机器人所处环境一般非常复杂、无法预知,单一传感器无法全面表达环境信息,所以多传感器信息融合技术的应用成为必然,是机器人解决避障问题的关键技术之一。然而自主避障问题在迎宾机器人开发过程中没有引起足够的重视,所以本文以实验室迎宾机器人为平台,对多传感器信息融合在迎宾机器人避障问题中的应用展开研究,有利于迎宾机器人的应用与推广。本文主要研究模糊逻辑信息融合和模糊神经网络信息融合算法在迎宾机器人避障问题中的应用,并对传统的T-S模型的模糊神经网络算法进行简化,减少神经网络训练时间,随后对上述算法在迎宾机器人避障问题上的可行性进行验证。首先,调研了多传感器信息融合技术在机器人中的应用,提出利用超声波传感器、红外传感器、电子罗盘和双目相机获取机器人运动环境信息的融合方案;对迎宾机器人进行运动学分析,建立运动模型。其次,对模糊逻辑信息融合进行研究,为机器人设计模糊控制器,在MATLAB平台下设计仿真实验,通过实验发现,模糊控制下迎宾机器人虽然能对环境进行识别完成避障任务,但是环境复杂时,容易出现控制规则匹配混乱,机器人左右摇摆问题,不利于迎宾机器人安全运行。然后,对传统的基于T-S模型的模糊神经网络信息融合进行研究,并设计仿真实验。通过实验发现,迎宾机器人避障稳定性得到提高、避障动作响应更加迅速,有效的解决了模糊控制存在的问题。最后,将改进的模糊神经网络算法移植到单片机上,开发上位机控制程序。通过机器人避障实验证明,本文改进的模糊神经网络算法,能有效解决机器人左右摆动问题,机器人避障稳定性得到极大提高。
[Abstract]:With the development of robot and artificial intelligence technology, intelligentization has become a topic of robot development. In order to realize intelligent development, it is necessary to solve the problem of autonomous obstacle avoidance, because the environment of welcoming robot is very complex. Unpredictable, single sensor can not fully express environmental information, so the application of multi-sensor information fusion technology is inevitable. It is one of the key technologies for robot to solve obstacle avoidance problem. However, the problem of autonomous obstacle avoidance has not attracted enough attention in the development of welcoming robot, so this paper takes the lab welcoming robot as the platform. The application of multi-sensor information fusion in obstacle avoidance of welcoming robot is studied. This paper mainly studies the application of fuzzy logic information fusion and fuzzy neural network information fusion algorithm in the obstacle avoidance problem of welcoming robot. The traditional fuzzy neural network algorithm of T-S model is simplified to reduce the training time of neural network. Then the feasibility of the above algorithm in obstacle avoidance of welcoming robot is verified. This paper investigates the application of multi-sensor information fusion technology in robot, and proposes a fusion scheme to obtain robot motion environment information by using ultrasonic sensor, infrared sensor, electronic compass and binocular camera. The kinematics analysis of the welcoming robot is carried out, and the motion model is established. Secondly, the fuzzy logic information fusion is studied, the fuzzy controller is designed for the robot, and the simulation experiment is designed under the MATLAB platform. Although the fuzzy control robot can recognize the environment and complete the obstacle avoidance task, when the environment is complex, it is easy to appear the chaos of control rules matching, the robot swinging from side to side, which is not conducive to the safe operation of the welcoming robot. The information fusion of traditional fuzzy neural network based on T-S model is studied, and the simulation experiment is designed. It is found that the stability of obstacle avoidance is improved and the response of obstacle avoidance is more rapid. Finally, the improved fuzzy neural network algorithm is transplanted to the single chip computer to develop the upper computer control program. Through the robot obstacle avoidance experiment, the improved fuzzy neural network algorithm is proved. It can effectively solve the problem of robot swinging from left to right, and the stability of obstacle avoidance is greatly improved.
【学位授予单位】：山东科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【参考文献】