Research On Binocular Vision System Of Heavy-duty Casting Robot

With the development of intelligent factories and industrial robots,heavy-duty pouring robots play an increasingly important role in the pouring industry,and their intelligence level is also increasing.However,the traditional heavy-duty pouring robot is still controlled by human directly,and the pouring workers need to be in a harsh pouring operation environment.Therefore,a binocular vision system for heavy-duty pouring robot is developed.Combining binocular vision system with heavy-duty pouring robot system,the heavy-duty pouring robot has the ability to perceive the surrounding environment,and can recognize and locate the target object according to the three-dimensional spatial information of the surrounding environment,thus realizing further artificial intelligence of heavy-duty pouring robot,and freeing the pouring workers completely from the harsh working environment.Firstly,the development of industrial robots,heavy-duty pouring robots and binocular vision systems is analyzed in detail.According to the research status of heavy-duty pouring robots and binocular vision systems at home and abroad,combined with the problems faced by heavy-duty pouring robots and their structural characteristics,the whole scheme of binocular vision for heavy-duty pouring robot is designed,and the experimental flow of digital image acquisition,camera calibration,image preprocessing,feature extraction and stereo matching,and three-dimensional reconstruction are determined.The software design of binocular vision system is completed.Then according to the principle of binocular camera imaging and the research of traditional binocular camera calibration method,the appropriate calibration method is selected for calibration experiment.Then,the method of image preprocessing is studied.Based on the study of the principle of classical SURF feature extraction algorithm,an improved SURF feature extraction and stereo matching algorithm is proposed.The feature descriptor is reduced from 64 to 26 dimensions by using the circular neighborhood of feature points,which greatly improves the operation speed of the algorithm.Then,the RANSAC algorithm is used to eliminate the mismatching points,which improves the accuracy of feature point matching.Finally,the principle of three-dimensional reconstruction of space points is deeplystudied.According to the main steps of three-dimensional reconstruction,the main flow of three-dimensional reconstruction is designed.The sand box is reconstructed by parallax principle.According to the depth image and three-dimensional information generated by the experiment,the accurate location of the sand box gate is realized.Based on the research of heavy-duty pouring robot and binocular vision system,the binocular vision system of heavy-duty pouring robot is designed.The algorithm of feature extraction and stereo matching based on SURF is improved.Finally,the simulation experiment platform based on binocular vision system is built.Experiments show that this research greatly improves the work efficiency of heavy-duty pouring robot,and also ensures the safety of life and property of pouring workers,which provides technical guidance for further putting it into practical production.Figure [72] table [6] reference [77].