Research On Location And Navigation System Of Transportation Robot Based On Radar Laser

With the vigorous development of e-commerce industry,warehousing logistics industry has also improved to a new direction of progress.Among them,replacing traditional manual transportation with mobile robots has become the mainstream trend of the development of warehousing logistics industry.Handling robots are becoming more and more popular in warehousing enterprises because of their high fatigue strength,low error rate and simple maintenance.At present,for the path navigation of transport robots,the most practical applications are magnetic bar navigation and two-dimensional code recognition navigation.The fixed manual instruction makes the robot become "traceable".However,when the scale of warehouse becomes larger and the scene of cargo placement changes rapidly,the problem of robot congestion and inability to reach the destination based on magnetic bars and two-dimensional codes is easy to cause.If the magnetic bars and a large number of two-dimensional codes are replaced,the cost of updating is very high.Therefore,the automatic navigation and transportation robot,which can choose the optimal path arbitrarily and make obstacle avoidance automatically and quickly,has become another research and development hotspot.Automatic navigation system mainly includes these aspects: 1)acquisition of environmental information;2)accurate positioning function;3)calculation of optimal path;4)release of robot control speed.At present,most of the transport robots developed to realize the functions of automatic path planning and obstacle avoidance are multi-sensor,expensive hardware and complex driving and control systems,so the production cost is high,which is not conducive to the application of enterprises in the market.Therefore,aiming at the characteristics of low cost,high reliability and high efficiency in the production of transportation robots by enterprises,this paper designs an automatic navigation transportation robot scheme based on ROS(Robot Operating System)from both hardware and software aspects,at the mean time studies and some improves are used on some algorithms.The main contents of this paper are as follows:1.Hardware system module.Embedded industrial computer and Arduino Pro are used as dual core processors.Among them,Arduino pro,as a lower computer,is responsible for recording the mileage information of the robot and receiving the speed commands issued by the upper computer to control the motor movement.Embedded industrial control computer assumes the function of upper computer,receives the measurement information obtained by radar laser,and carries out a series of calculation of navigation core algorithm.2.Optimal path calculation module.Dijkstra and A* algorithm are compared,and a new potential layer is set up to make the planning path more reasonable.3.Based on the improved particle filter positioning module.In this paper,the measurement information based on radar laser is fused with the mileage information acquired by motor codes,and the particle sampling update phase of particle filter is improved to achieve faster convergence effect.In addition,the motion state of the robot is classified by using the neural network,so that the sampling and updating of the particles can be adapted to the actual trajectory.4.Speed release module.In the framework of Dynamic Window Approach(DWA),the evaluation function in DWA is mainly improved in view of the left-right swing phenomenon in the actual operation of the robot.Finally,through the implementation of the self-designed transport robot entity,the experimental results show that this module can run normally,and some of the algorithms are improved by obvious effect compared with traditional algorithms.Although the navigation system of transport robot designed in this paper has only two sensors strictly,and the cost of consumables and processors is low,it runs well,which has certain guiding significance for enterprises to develop low-cost,reliable and efficient transport robots.