Font Size: a A A

Real-time Study And Analysis Of Gateway Based On CAN-CANFD Network

Posted on:2020-11-10Degree:MasterType:Thesis
Country:ChinaCandidate:X S ChengFull Text:PDF
GTID:2392330590971779Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
With the increasing of automobile functions,the number of messages that need to interact on the whole network increases exponentially.When using the traditional CAN bus to communicate on-board network,it is limited by bandwidth and communication rate.Therefore,the communication of a large number of messages may cause the blockage of the whole network.In serious cases,the loss of some messages may cause vehicle safety problems.In order to improve these limitations faced by traditional CAN network,CANFD bus is used to alleviate these problems.When multiple buses are used in the network,the scheduling of gateway is very important.If some messages are scheduled to be delayed,they will be discarded due to the delay.Therefore,it is of great significance to study the scheduling of CAN-CANFD network gateway for the development of automobile technology.Thesis mainly uses the gateway in CAN-CANFD network as the research object,analyzes the delay problem of gateway scheduling in this network,optimizes the existing gateway scheduling algorithm,proposes the optimized scheduling algorithm,and finally the CAN-CANFD network and The data was designed and simulated on the hardware-in-the-loop simulation experimental platform,and the feasibility of the optimized scheduling method was verified.The main contents of this paper are as follows:1.Firstly,the current problems of CAN-CANFD network gateway are analyzed,and the transmission problem is studied.On this basis,the current mainstream gateway transmission scheduling algorithm is analyzed,and the shortcomings of the current mainstream scheduling algorithms are analyzed.The scheduling algorithm of this thesis is proposed.2.Secondly,for the WRR scheduling algorithm,when the low-weight queue is scheduled for polling,it will be scheduled at the end of the polling period.When applying the algorithm,the scheduling of long-period packets may be delayed.Therefore,the WRR is improved for this problem,and the scheduling order is smoothed for one polling period.The simulation results show that the smoothed scheduling sequence can improve the delay caused by the scheduling of a large number of high priority short-period messages.Furthermore,in order to prevent high priority messages from being suppressed by low priority messages in the periodic queue and to speed up data processing,this paper uses ID priority queuing to schedule,and uses the combination of data and linked list to store CAN messages in the data storage mode.The simulation results show that the combination of data and linked list can effectively reduce the processing time of messages.3.Then the CAN-CANFD network topology is designed,and the ECU node message data in each network segment is designed,including the message,message ID,message period and each ECU node that each ECU needs to send.Message interaction routing.4.Finally,in order to verify the proposed gateway scheduling algorithm and the feasibility of the designed CAN-CANFD network,a CAN-CANFD network is built in the CANoe simulation environment,and a gateway based on the embedded platform is developed to form a hardware-in-the-loop simulation experimental platform.Segments to CANFD network segment packets for forwarding testing,and testing the performance of the gateway.The verification test results of the algorithm show that the optimized scheduling algorithm can effectively forward the packets from CAN segment to CANFD network segment,which proves the method and the feasibility of the network.It is real-time and stable through the network.The test results show that the scheduling algorithm used in this paper can effectively reduce the delay time of long-period packets.When the network load rate is higher,the improvement is more obvious,up to 21.78%.
Keywords/Search Tags:CAN-CANFD network, gateway, delay, scheduling algorithm
PDF Full Text Request
Related items