An AGV obstacle avoidance method based on lidar and depth camera data fusion

Abstract

The invention discloses an AGV obstacle avoidance method based on lidar and depth camera data fusion, comprising: a controller; a data processing unit, a decision-making unit and a control unit are arranged in the controller; and a laser installed at the front end of the AGV Radar and depth cameras, and ordinary cameras installed on the AGV; the lidar, depth cameras, and ordinary cameras are respectively connected to the controller signal for signal transmission. In the present invention, the combination of laser radar data and depth camera data is used to scan and track obstacles in the direction of AGV travel. By preprocessing the sensor data, a corresponding data node is established in the system for data transmission. The controller and Nodes communicate to obtain different data. The data processing unit is used to process sensor data and run algorithms for data processing. The processing results are determined by the decision-making unit to determine how the AGV avoids the current situation, and the avoidance method is transmitted to the underlying motion control unit for Complete evasive maneuvers.

Description

technical field [0001] The invention relates to an automatic drug distribution system, in particular to an AGV obstacle avoidance method based on laser radar and depth camera data fusion. Background technique [0002] As an emerging technology that has emerged in recent decades, intelligent robots and their automation technology have continued to grow and enter people's lives. In order to make robots better serve human life, the development of robots is slowly moving from the workshop assembly line From assembly robots to indoor sweeping robots, ordering robots in restaurants, etc. Robotics is increasingly becoming an integral part of human life. [0003] However, the existing obstacle avoidance solutions for intelligent robots have the defect of singleness, which makes the robot unable to cope with complex environments well, and the collected data does not have enough information to deal with obstacle avoidance operations. If multi-line laser radar is considered, although...

AI Technical Summary

Problems solved by technology

[0003] However, the existing obstacle avoidance schemes of intelligent robots have the defect of singleness, which makes the robot unable to cope with the complex environment well, and the collected data does not have enough information to deal with the obstacle avoidance operation

If you consider using multi-line laser radar, although the data contains multi-dimensional information, it will help improve the adaptability of the robot, but the hardware cost will be greatly increased.

Images