摘要
为了提高机器人避障地图的准确性和避障算法的效率,本文对VFH+算法进行改进。将超声波传感器分组分段进行工作以避免相互之间的干扰,对超声波传感器返回的数据进行软件滤波以增强实时地图的准确性;改进代价判定函数以便对通行代价进行更准确的标识;设计基于动态阈值的局部循环跳出机制以克服固定阈值导致的问题。仿真结果表明改进的算法是有效的。
In order to increase the accuracy of robot avoidance map and the efficiency of robot avoidance method, this paper im- proves the VFH ＋ method. First, we group the ultrasonic sensor to avoid interference between each other. Second, we use soft- ware to filter statistics returned from ultrasonic sensors to increase the accuracy of real-time map. Meanwhile, we improve the cost function to get more accurate identification of cost. Lastly, we design a mechanism that allows the robot to get out of the loop based on dynamic threshold. Simulation shows that the improvements are effective.
出处
《计算机与现代化》
2014年第4期97-101,104,共6页
Computer and Modernization
基金
国家科技重大专项经费资助项目(2013ZX040060011-102-002)
关键词
机器人
避障
地图创建
软件滤波
超声波传感器
robot
obstacle avoidance
map constitution
software filter
ultrasonic sensor
