Research And Improvement Of Collision Detection Algorithm Based On Virtual Proximity Sensor

In the era of Industry 4.0,digital twins have become a winning weapon for many companies to transform to digital and intelligent.Its core is to virtualize all links in the actual production and manufacturing into a digital platform through multiple digital means,forming a mirror image of virtual manufacturing and actual manufacturing.Therefore,the concept of digital twins plays a vital role in optimizing the entire enterprise process and enhancing the competitiveness of the enterprise.Especially in the intelligent manufacturing industry,it is bound to promote the rapid development of intelligent manufacturing.However,few domestic companies currently apply the concept of digital twins to proximity sensors.Especially in the industrial assembly line,the collision detection algorithm of the proximity sensor still uses the SOLID algorithm,and the detection efficiency of the algorithm is low,so that the industrial production will cause a decrease in productivity due to the occurrence of collisions.Therefore,there is an urgent need for a method that can tune the proximity sensor and improve the efficiency of collision detection to help more companies move toward digitalization and intelligence.In order to be able to tune the proximity sensor,this thesis starts with virtual simulation and generates a virtual proximity sensor.The virtual proximity sensor is tuned through the digital platform to meet the needs in real production.Then we research the collision detection algorithm based on the optimized proximity sensor.For the collision detection algorithm,this thesis selects the hierarchical bounding box algorithm for research.The hierarchical bounding box algorithm can quickly detect that two objects have not collided,so it is widely used in games and channel communications.This thesis focuses on the in-depth research on the bounding box collision detection algorithm.The specific research results are as follows:(1)Conduct related experiments on common bounding box algorithms and get comparative results.Analyzing the main factors affecting the efficiency of collision detection,based on these factors,an Axis-Aligned bounding box collision detection algorithm based on B+ tree storage is proposed.The hierarchical bounding box formed by the object is divided into multiple bounding boxes by the split plane theorem and stored in each node of the B+ tree.Due to the special nature of the B+ tree,its non-leaf nodes only store the index,and the specific values are only stored in the leaf nodes,which greatly reduces the memory space occupied by the entire hierarchical bounding box tree and speeds up the execution efficiency of the algorithm.(2)This thesis proposes an optimized intersection test algorithm between triangles,and divides the intersection test into two cases: the same plane and different planes.The positional relationship between the same two triangular faces is different,then the intersection test will be completed in three-dimensional space;on the contrary,it can be projected into two-dimensional space and judge whether they intersect through simple geometric positional relationship.In this way,not only the robustness of the algorithm is enhanced,but the efficiency of collision detection is also improved.Experimental data shows that the optimized algorithm has a significant improvement in the efficiency of collision detection.Applying this algorithm to NX simulation has great value and significance for collision detection in intelligent manufacturing.