Four Main Coke Oven Vehicles Unmanned Hybrid Enhanced Intelligent Active Safety Research

Railcars have a large load and are mostly used to transport large-quality items.They have an irreplaceable role in long-distance transportation and large-scale industrial production.The working environment of the railcar is harsh,and it is easy to collide with surrounding objects during operation,and there is a huge hidden danger.Once a collision accident occurs on a rail car,it will not only destroy its vehicle structure and cause casualties,but also seriously affect the efficiency of industrial production and transportation and cause huge economic losses.Therefore,solving the collision safety problem of rail vehicles is of great significance to ensure the safety of rail vehicles and improve the efficiency of industrial production.At present,the main research direction for preventing collision accidents is to improve the active safety performance of vehicles,but there are few active safety methods specifically developed for rail vehicles.Most of them are researched for road vehicles.However,the running speed,working environment and working production method of rail cars are different from those of automobiles.Existing active safety algorithms are difficult to apply directly.For rail vehicles,the development of its active safety algorithm enables it to identify possible collision risks as early as possible,and accurately predict the direction and position of the next moment of movement of the moving object in front of the rail car,so that it can deal with "coming" collision Early warning and control of risks to improve the active safety performance of rail cars.How to predict the trajectory of the moving objects in the vehicle scene through the parameters and algorithms of the existing vehicle running on the road is an important issue that must be considered in the design of the active safety system of the railcar.Based on the idea of active safety design,this article takes Bao Gang’s four main coke oven vehicles as the research object,and uses different UWB positioning systems to build railcars to run rail base stations to detect regular moving objects(fixed workers)for different types of rail road environments.Using a 16-wire laser radar scanning to detect the motion trajectory of irregular moving objects(random pedestrians and vehicles),a real experiment was carried out in Bao Gang’s coke oven area to obtain a large number of real parameters of the moving trajectory of moving objects.Taking into account the different working nature and road environment of the four main coke oven vehicles,predicting the trajectory of various moving objects is the next moving direction.The main research contents and innovations of this article are summarized as follows:(1)Aiming at the situation where only fixed staff members pass through the coal truck operating area,a hybrid trajectory prediction method based on the gray Markov model is proposed for the four main coke oven vehicles hybrid enhanced intelligent trajectory,which collects data on the pedestrian walking position in the operating area of the coke oven coal truck According to the hybrid enhanced intelligence method of people in the loop,classify and upload regular moving objects based on human cognition.Based on this,a Markov model of pedestrian trajectories in the coal truck area is established in combination with the gray prediction method,and the characteristics of the trajectory changes during walking are identified.Compared with the results predicted by the traditional gray model,the error is smaller and the accuracy is higher.It can be applied to the unmanned operation system of the coke oven coal car to realize the unmanned active safety control of the four main coke oven vehicles.(2)Aiming at the situation that there are random pedestrians and various vehicles in the large lane of the coke pushing vehicle,a hybrid enhanced intelligent trajectory prediction method based on DEGWO-SVM is proposed.The hybrid enhanced intelligence method of human-in-the-loop is used to identify and classify types of irregular moving targets in the coke oven area,so that the next prediction of each type of moving target can be accurately performed,and then the differential evolution mechanism(Differential Evolution,DE)and the gray wolf algorithm are combined.(Grey Wolf Optimization,GWO)Optimize the kernel function parameters and penalty factors of the Support Vector Machine(SVM)to reduce the influence of the harsh environment on the overall prediction model operation,obtain the optimal solution of the SVM parameters,and establish Accurately predict the trajectory of the moving object in the harsh environment of the lane of the coke pusher,thereby controlling the operation of the coke pusher to achieve active safety.(3)An improved hybrid grey wolf(HGWO)optimized SVM hybrid enhanced intelligent prediction algorithm is proposed.Based on the DEGWO-SVM hybrid enhanced intelligent prediction algorithm,the differential evolution mechanism was improved and adjusted adaptively,and the SVM kernel function parameters and penalty factor C were combined with levy flight optimization build an HGWO-SVM hybrid enhanced intelligent prediction model based on adaptive difference and Levy Flight Improvement,and perform simulation and field experiments respectively And compared with DEGWO-SVM hybrid enhanced intelligent prediction results to verify the accuracy and superiority of the method.