Active anti-collision detection method based on material piling and taking real-time operation dynamics

• Main Authors: GAO JIN, CAI XINGEN, XU LIYUN, WANG MIHUAN, ZHAO XIANGQIAN, LUO WEIQIANG, CHEN LING, YANG HANFANG, YANG DUOBING, CUI JIABIN, SUN WEIZHE
• Format: Patent
• Language: Chinese; English
• Published: 04.06.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; COUNTING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PHYSICS; REGULATING

The invention relates to an active anti-collision detection method based on material piling and taking real-time operation dynamics. The active anti-collision detection method comprises the following steps: acquiring equipment attitude and state information; the collected PLC data are analyzed and judged; starting a target detection module to analyze an object in a video stream of a camera on the equipment; and the target detection module outputs a result and then processes the result, sets three areas according to the visual angle of the video picture: a normal area, a warning area and an emergency stop area, and performs further judgment when the frame of the target object intersects with the warning area or the emergency stop area and the confidence is higher than a preset threshold. According to the PLC control system and the target detection module based on deep learning, collision between the stacker-reclaimer and floating personnel and vehicles in the operation process can be effectively prevented. The