A pixel-level target location method based on fusion of laser and monocular vision

Abstract

The present invention relates to the field of object three-dimensional positioning methods, in particular to a pixel-level target positioning method based on the fusion of laser and monocular vision. The relative position is calibrated to obtain the calibration parameters. After the monocular camera is used to obtain the environmental image, preprocessing is performed and the pixel coordinates of the target to be located are determined. Then, the actuator is driven by the PID feedback control to drive the laser ranging module to rotate to collect data and perform calculations. Finally, the laser ranging module is driven to accurately hit the ranging light spot to the target position, and then the data is collected to achieve target positioning through calculation. The invention consumes low computing resources, can ensure real-time performance under the condition of limited mobile terminal computing power, has high-performance object positioning capabilities, high precision, and wide range, can improve the operating range of intelligent autonomous operation equipment, and has a certain cost Advantages, lower the threshold of use, and facilitate technology promotion and application popularization.

Description

technical field [0001] The invention relates to the field of three-dimensional object positioning methods, in particular to a pixel-level target positioning method based on laser and monocular vision fusion. Background technique [0002] In recent years, with the development of robot technology and artificial intelligence technology, more and more intelligent autonomous operation equipment has been applied in the fields of social production and life, military and national defense. In social production and life, it is a trend of social development to use intelligent robots to assist or replace humans in some repetitive or service work, which can effectively improve production efficiency and improve the convenience of daily life; in the field of military defense, the use of intelligent robots To assist or replace people to perform specific tasks can effectively improve the efficiency of task execution and ensure the safety of task performers. [0003] Intelligent autonomous o...

AI Technical Summary

Problems solved by technology

At present, depth sensors such as binocular vision, laser radar, and millimeter-wave radar are usually used in the perception system for object positioning. There are still obvious deficiencies in positioning accuracy, effective positioning distance and cost

Since intelligent autonomous operation equipment is constrained by issues such as cost, volume, and power consumption, the computing power of the terminal computing equipment it carries is relatively limited. Therefore, the object positioning method using the above scheme is expensive, has poor real-time performance, and has limited positioning range. It is difficult to effectively complete the precise positioning of objects under the actual working conditions of the equipment

Images