Automatic identification striking system and method for robot shooting confrontation match

Abstract

The invention discloses an automatic identification striking system and method for a robot shooting confrontation match. The system comprises an enemy robot, an our robot, a front camera, a rear camera, a holder and a gun barrel. According to the automatic identification method, a self-adaptive traditional algorithm is adopted, light bars are detected by extracting color features and brightness features of the light bars on the two sides of the armor plate of the enemy robot, and the armor plate of the enemy robot is detected through the positions of the light bars and multiple logic conditions, so that the enemy robot moving at a high speed is identified; the numbers on the armor plate of the enemy robot are identified by extracting edge information and distinguishing the numbers by a classifier, so that necessary enemy robot information is provided for a decision algorithm; and in automatic aiming and striking process, depth information is provided through an infrared camera to obtain the relative position of the enemy robot and the our robot, and the robot holder is controlled to track the enemy robot, so that automatic aiming is achieved, state prediction of the enemy robot isadded, and advanced pre-judgment and accurate striking are achieved.

Description

technical field [0001] The present invention relates to the technical field of robots, in particular to an automatic identification strike system and method for robot shooting confrontation competitions. Background technique [0002] With the development of artificial intelligence technology, related research in the field of robotics has become a hot topic today. Robots can assist or replace humans in repetitive and dangerous tasks, bringing great convenience to people's lives. At present, related technologies in the field of robotics include environmental perception, autonomous positioning and navigation, path planning and intelligent decision-making, etc. [0003] Robot shooting competition is an important form of competition in the field of robotics, such as the RoboMaster Artificial Intelligence Challenge. The competition is an international competition jointly sponsored by DJI and the Global Conference on Robotics and Automation (IEEE International Conference on Robot...

AI Technical Summary

Problems solved by technology

Since both robots are in high-speed motion during the competition, the camera shake of the own robot and the irregular high-speed motion of the enemy robot will lead to blurred camera images. How to accurately identify the enemy robot under this condition and achieve precise aiming and strike has become a major technology. problem

[0005] In automatic recognition, the common traditional method achieves tasks such as target recognition through feature extraction or template matching. It can achieve high accuracy in a specific environment, but it cannot be applied to a variety of environments and has poor robustness.

Deep learning has a wide range of applications in visual tasks such as target recognition and deblurring, and has high robustness. However, the calculation cost of deep learning is high, which is limited by the performance of the mini-computer used by the competition robot, and it is impossible to achieve a relatively high performance. High frame rate recognition

[0006] Automatic aiming needs to obtain the depth information of the enemy robot. The depth estimation method combined with the monocular camera and the PNP algorithm used in the past has a large error and cannot achieve accurate ranging.

Images