Research On Bionic Path Planning Algorithm For Mobile Welding Robot

With the rapid growth of technology,the use of robotics is wide.Mobile welding robots,as one of the vital parts of robotics,integrate mobile with welding technology and have a wide range of prospects.Path planning is the core of mobile technology,and the purpose is to arrive at the end position safely and at the fastest speed.Nowadays,the complexity places higher demands on adaptation and optimization of planning methods.The problem of long distance,many points and long time was addressed by a complex method based on Ant Colony Optimization.The study focused in three settings:1)To address its vulnerability to U-traps and low route quality in simple environments,the successful ant colony algorithm was proposed.Deadlock information is stored as scenario memory.Incorporate the endpoint factor in the heuristic information.Update the total pheromone using the metrics of successful ants.The algorithm was1.84% shorter in length and 8 fewer inflection points compared to the classical method.2)For its slow planning speed in complex context,a successful ant colony algorithm based on the nesting mechanism is proposed.The initial pheromone is assigned using a cognitive degree nonlinearity,and the added pheromone values are controlled using a modulation factor.The planning efficiency of the successful ant colony algorithm was improved,with a length reduction of 5.64% and a reduction in inflection points by 4.3)For its collision-prone problem in dynamic environment,a successful ant colony dynamic planning algorithm based on rolling window and nesting mechanism was proposed.A successful ant colony algorithm using nesting was used to obtain the global path,and when it moved,the map was updated and collisions were predicted using a scrolling window.Then local replanning was performed according to the three planning schemes.The results showed that the method was able to accomplish dynamic planning.In sum,these methods can realize its path planning in different contexts respectively,providing methods for real usage.New ideas are provided for storage and patrol robots.Figure 52;Table 2;Reference 58...