Font Size: a A A

CAN Network Management Based On OSEK/VDX Specification

Posted on:2011-04-19Degree:MasterType:Thesis
Country:ChinaCandidate:Y H ChenFull Text:PDF
GTID:2132360305955321Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Along with the booming of the integrated electronic techonolgy, more and more electronic devices which use the electronic control units(ECUs) as cores are introduced into the power system and the control system, such as the airbags, central lockings, and the power windows etc. The application fields and the functions of these devices have already becoming the judging measurement for the performance of the vehicle to a large extent. However, some problem also has come up, such as the complex structure of wiring harness, the lack of available space and the less reliability. In order to simplify the complicated wire harnesses, and raise the data transmission ratio between the ECUs at the same time, the techonolgy of in-vehicle bus emerges as the times reqire. This technology decreases the fault rate of the wire harnesses enormously, and is convenient for the maintenance and overhaul.CAN, LIN, MOST, FlexRay, are the mainly dominating in-vhiecle bus protocols until now. They can be used for different purposes according to the data ratio and the transmission quality required by the applications in vhiecle. The protocol in CAN bus is an in-vehicle communication protocol which is used widely for controlling the electronic devices and power system. It has two different data transmission ratio: for the low speed CAN it is 100kbps, for the high speed CAN is 1Mbps. In order to decrease the number of the wire harnesses, and raise the data rate, CAN makes some optimal designs for the in-vehicle serial bus. It proceeds from actual conditions using the construction of multi-tasks serial, and also has some capability in fault tolerance. All the dominating manufactures over the world have already been using CAN as a standard protocol in vehicle now days.The system of the in-vehicle bus connects the ECUs which have requires for signal transmission or data exchange by two cables. Each ECU could be seen as an individual node, and an in-vehicle network can be upbuilt by connecting some individual nodes together. This kind of network is similar with the common PC network in structure. But the most difference between them is that the in-vehicle network is designed based on the distributed system which in order to avoid the overload of the network, otherwise the in-vehicle applications have strict reqiures on the safety and the reliablity of the vehicles. Consequently, it expects that the network management system based on the direct NM of OSEK/VDX specification can be introduced into the in-vehicle CAN network to ensure the reliability and the robustness of the in-vehicle network.There are two different types of message exchange communication in the in-vehicle laboratorial CAN network: the exchange which upbuilds and maintians the NM communication in the logical ring, and the exchange of application message communication between CAN node and the in-vehicle gateway. The in-vehicle gateway implements the data exchange with the remote controlling server through the IPv6 wireless network. The extent frame which has 29bits extent identification is used as the standard message format, and the in-vehicle communication protocol is defined based on it. Due to the the NM messages and the application messages have different communication objects, besides the node ID, group ID is assigned to each node. Both of them give each node a unique identification in in-vehicle network. Moreover, in-vehicle gateway and the CAN nodes accept the messages transmitted on bus selectively by the different group ID, and so avoid the processing of a mass of simmilar unrelated messages.The works refered in this dissertation based on the National Development and Reform Commission (NDRC) project"IPv6-based in-vhiecle information system research and industrilaization". These works provide certian technical support to the developing of the next generation of in-vhiecle information system.During the processing of implement, two basic internal states NMShutDown and NMOff are omitted from the internal states of the direct NM at first. As soon as the node electrified, it transfers into the only basic state NMOn. And then, the substates NMInit and NMAwake have been reserved, while the substate NMBusSleep and the paratactic NMActive and NMPassive have be omitted. Eventually, all the substates NMReset, NMNormal and NMLimphome have been resered, except for some functions in NMLimpHome.Ususally, after the startup of the in-vehicle laboratorial CAN network, nodes participate into the logical ring randomly through the NM communication. Due to the uncertainty of the factors which determine the opportunity to participate in the logical ring, the phenomenon and the result of the experiments are hard to analyse in this case. Consequently, the nodes in network should participate into the logical ring in order according to the ID from small to large. During the upbuilding process of the stable NM communication, the mechanism of how to determine the logical successor, being skipped in the logical ring and the usable method of timer TTyp and TMax can be proved that they have be carried out with all the functions and behaviors described in the direct NM of the OSEK/VDX specification, by the message content of the in-vehicle network got from the USBCANII, and the results from the testing codes shown by the LED s in the testing circuit. However, in the case of some nodes in network have turned absent, the former experiment project would make an unexpected counteraction in the rebuilding process of the stable NM communication. It actually destories the euqal status of the nodes when they participate into the logical ring randomly, and counteracts the rebuilding of the stable NM communication; because of the former experiment projcet has made the nodes get online in order. Therefore timing function and relative processing mechanism for the configuration imformation need to be introduced in, to verify that all the online nodes could participate into the logical ring rebuilding the stable NM communication again in time, after any node's absence.This paper discusses how to design a network management system based on the theory of the direct NM of OSEK/VDX specification, and the process of its application in the in-vehicle CAN network. It's the combination of the advanced theory network management and the practical technology in-vehicle bus. All the content in this paper has been supported by related projects.
Keywords/Search Tags:CAN bus, direct NM, OSEK/VDX, in-vehicle network management
PDF Full Text Request
Related items