融合嵌入式视觉与激光测距的机器人自主避障方法  

Robot Autonomous Obstacle Avoidance Method Based on Embedded Vision and Laser Ranging

作　　者：周晖 赵玲峰 ZHOU Hui;ZHAO Lingfeng(Guangxi Agricultural Vocational University,Nanning 530007,China)

机构地区：[1]广西农业职业技术大学,南宁530007

出　　处：《机械设计与研究》2023年第2期26-29,共4页Machine Design And Research

基　　金：广西高校中青年教师科研基础能力提升项目(2019KY0975,2022KY1214);高功率激光应用实验室建设项目(2020YT05G087)。

摘　　要：多数智能服务机器人处于敞开式环境中,在复杂部署工作环境中障碍物的位置及形状均具有较强的随机性,导致机器人为完成避障出现路径重复、收敛速度偏慢的问题。为此,提出融合嵌入式视觉与激光测距的机器人自主避障方法。构建激光测距和嵌入式视觉传感器模型,利用尺度不变特征变换算法,提取并融合数据特征;当机器人感知到前方有障碍物时,首先考虑左侧通行,若机器人位置靠右,则根据最优路径选择右侧通行;基于通行规则设定合适的转向角度,保证机器人成功躲避障碍物。实验测试结果证明,所提方法可精准感知机器人周围环境中的障碍物,且避障路径合理,收敛速度更快。Most intelligent service robots operate in an open environment.The location and shape of obstacles in a complex deployment work environment have strong randomness,which leads to the problem of repeated paths and slow convergence speed for robots to complete obstacle avoidance.Therefore,a robot autonomous obstacle avoidance method based on embedded vision and laser ranging is proposed.The model of laser range finder and embedded visual sensor is constructed,and the data features are extracted and fused using the scale invariant feature transformation algorithm.When the robot senses that there is an obstacle in front,it first considers the left side to pass.If the robot is located on the right side,it chooses the right side to pass according to the optimal path.The appropriate turning angle is set based on traffic rules to ensure that the robot can successfully avoid obstacles.The experimental results show that the proposed method can accurately perceive the obstacles in the robot's surrounding environment,and the obstacle avoidance path is reasonable and the convergence speed is faster.

关 键 词：嵌入式视觉传感器 激光测距传感器 自主避障 最优路径 转向角度 

分 类 号：TP245[自动化与计算机技术—检测技术与自动化装置]