Research On Optimization Method Of Fire Fighting Path For Robots With Multiple Ignition Points

Today,fire is one of the main disasters that most often and commonly threatens public safety and social development.It is a common catastrophic problem faced by people all over the world.It has caused many lives and properties to human society.serious losses,a serious threat to public safety.Therefore,how to carry out fire fighting at the fire scene safely and quickly in order to minimize the losses when the fire has occurred has been a difficult problem that has plagued the society for a long time.With the advancement of the times,informatization and intelligence have developed in all walks of life,and the same is true in the field of fire rescue.In the actual fire,the high temperature,toxic and harmful gases,and thick smoke that affect the line of sight,etc.,make emergency personnel often unable to go deep into the scene as soon as possible on the premise of ensuring personal safety,thus delaying the best time for emergency fire fighting operations..The use of robots to assist or replace manual firefighting and rescue can achieve twice the result with half the effort.The robot has the characteristics of high temperature resistance and smoke resistance.It can use the robot to go deep into the fire field and quickly and accurately determine the location of the ignition point,which is very important for formulating rescue plans and fire fighting strategies.The premise that the robot can complete the task autonomously and efficiently is to optimize a safe and fast path for it.Therefore,this paper mainly explores how to optimize the fire-fighting path for the robot to go deep into the fire field,and plan a reasonable fire-fighting obstacle avoidance path and a fire-fighting sequence with multiple fire points.,reduce social losses more efficiently and safely.The main work of this paper is as follows:(1)Explore the multi-firing point fire-fighting path optimization problem of robots is essentially a single-objective or multi-objective traveling salesman problem.Therefore,an improved ant colony algorithm and a dynamic multi-swarm particle swarm optimization algorithm are proposed for solving the traveling salesman problem and the multiobjective optimization problem,respectively.Do corresponding simulation experiments to verify their optimized performance.(2)Aiming at the optimization problem of firefighting paths for robots with multiple ignition points,firstly,the experimental analysis is carried out in the scene of multiple ignition points of a single target and single robot,the environment modeling of multiple ignition points is carried out and a mathematical model is established,and the improved ant colony algorithm and the basic ant colony are used respectively.Algorithm and beetle algorithm solve this problem,and obtain the optimal firefighting path scheme that only considers the cost of path distance for a single robot.Then the system realization of single-robot fire-fighting path is completed.After setting the optimization parameters,the system solves the problem and can quickly provide the optimal single-target single-robot fire-fighting path scheme with multiple ignition points.(3)Considering the dynamic changes of ignition potential,property value and distance cost of each fire,a mathematical model of multiobjective and multi-robot firefighting path optimization is established.First,the fire detection model trained by the YoloV5 framework is used to detect the ignition point of the fire field.Then,according to the different attribute data detected by each ignition point,dynamic simulation is carried out in fire simulation software,and multi-objective optimization is carried out combining property value and considering distance cost.The dynamic multi-swarm particle swarm algorithm,NSGA and NPGA are used to solve the problem respectively,and the optimal multi-objective and multi-robot firefighting path scheme is obtained.The main innovations of this paper are as follows:(1)An improved ant colony algorithm is proposed,combined with the beetle antenna search mechanism of the beetle algorithm,which improves the optimization performance of the ant colony algorithm.(2)Combined with the dynamic multi-swarm strategy,the particle swarm algorithm is improved,and the random reorganization strategy is introduced to improve the diversity of solutions in the solution process and speed up the convergence speed.(3)In the multi-objective optimization of the fire-fighting path of the robot with multiple ignition points,the YoloV5 target detection framework and the fire simulation software are combined to simulate the fire and combine it into the multi-objective optimization.