Study On Localization And Mapping Algorithm Of High-Power Intrinsically Safe Coal Mine Rescue Robot

When a gas explosion such as a gas explosion occurs in a coal mine,the coal mine rescue robot can replace the emergency rescue personnel to enter the underground disaster occurrence area for the first time,perform detection and inspection of the disaster area,and rescue the trapped miners.The participation of coal mine rescue robots has greatly increased the chances of successful rescues,while also reducing the casualties of rescue workers.However,most of the underground rescue robots are explosion-proof explosion-proof.If the explosion-proof casing is broken,it will cause a secondary explosion in the underground,causing greater losses and injuries.At the same time,the underground environment after the disaster has changed.The original map has been unable to reflect the downhole situation during the rescue process.The robot's downhole positioning and mapping capabilities are extremely important.This paper focuses on the three key technologies of coal mine rescue robot modeling and simulation,robot high-power intrinsically safe driving technology,robot underground real-time mapping and positioning algorithm.In this paper,the kinematics model is established from the physical quantities such as displacement,velocity and angular velocity of coal mine rescue robots.In order to judge the position and attitude of the robot in the underground in real time,based on the kinematics model,a sliding mode variable structure controller is designed to track the movement of the robot.Dynamic modeling and analysis are carried out for the special running attitude of the robot over the steep slope,which provides a theoretical basis for the matching of the driving force of the robot.In the process of research and development of robots,there are problems such as long development cycle,high risk,and difficult to correct in time.The simulation analysis of the robot is carried out by establishing a simulation test system for coal mine robots,which verifies the feasibility of robot design.Aiming at the existing explosion-proof coal mine rescue robot's hard injury--if the explosion-proof casing is broken in the falling rock,it is easy to cause the secondary explosion of the surrounding explosive gas,and it is proposed to use the intrinsically safe explosion-proof instead of the explosion-proof type.method.In order to solve the problem that the intrinsically safe robot can not achieve high power drive,the principle of "divide power integration" is proposed.Firstly,the correctness of the theory is proved by a series of hydrodynamic relations.Based on this theory,the method of multi-pump confluence is proposed,and the feasibility of multi-pump confluence is proved by software simulation.In order to optimize the control of the system,the mathematical model of the multi-pump merge system was established and the control method of PID control was established.At the same time,in order to allow the robot to travel a farther distance with limited power,a control method based on distributed control has been established.In order to verify the actual efficiency of multi-pump confluence,a hydraulic confluence test platform was designed.In order to make the multi-pump merge drive system meet the coal mine underground "ia" level requirements,the motor and battery pack were intrinsically modified,and finally the multi-pump merge hardware system was designed.Aiming at the accuracy of the coal mine rescue robot's downhole positioning and mapping algorithm,which can only achieve the first-order Taylor accuracy,a high-precision second-order central differential particle filter(SOCDPF)SLAM algorithm(SOSLAM)is proposed.Secondly,the robustness of FastSLAM and UFastSLAM algorithm is very easy to be affected by the noise caused by the harsh terrain in coal mine operation.A strong tracking adaptive UFastSLAM algorithm based on suboptimal fading factor is proposed.Finally,combined with the strong robustness,adaptability and high nonlinear approximation of second-order CDKF,and the addition of particle filters not affected by non-Gaussian and nonlinear models,a strong tracking second-order central difference SLAM algorithm is proposed.(STSOSLAM).The superiority of the above algorithms was verified by simulation experiments and construction of the office.Finally,in the actual coal mine ventilation test roadway,a real underground environment map was drawn.In order to solve the problem of moving the coal mine underground rescue robot from one point to another,the path planning algorithm based on probability road map(PRM)is studied.Through the simulation test,the influence of the particle number on the algorithm is verified,and the optimal particle number of the algorithm is calculated.