Vol.11 no.3 2019

Commentary : Contributing to the SpaceWire international standard (H. HIHARA et al.)−148−Synthesiology - English edition Vol.11 No.3 (2018) experiences for the development of data handling systems such as scientific satellites, practical satellites, and space stations that have been developed by Japan.3.1 Difference in viewpoint for optimal designThe Japanese proprietary specifications were utilized in the SpaceWire remote memory access protocol (SpaceWire RMAP), one of protocols in the SpaceWire international standard. RMAP is the protocol to read and write memories and others in the equipment connected to a network. Looking back at this process, it was found that there were two advantages to the Japanese development process.One was the skill in obtaining consensus by smoothing communication among the organizations that carry out the R&D. In the communication standard layer that was initially being overseen by the SpaceWire Working Group Committee (SpW WG) that was set in the European Space Research and Technology Centre (ESTEC), a research institution of ESA, there was a layer added to realize real-time properties. It was called SpaceWire-RT or SpaceWire-T. In this proposal, the interface that directly linked to the uppermost telemetry command layer had a complex specification, and an agreement could not be reached for nearly a year because of heated discussions. From our experience of development and operation of the data handling system for Japanese scientific satellites, this protocol layer had a heavy implementation load and was not practical. Those of us participating in the SpW WG through the Japan Aerospace Exploration Agency / the Institute of Space and Astronautical Science (JAXA/ISAS) realized one point. Europe is a society that defines the content of one’s job very precisely. The protocol layers of the communication standard had clear interfaces and could be easily divided into tasks, or enabled implementing specications in parallel, and perhaps that was the reason there were overlaps in several portions in the layers. In contrast, the implementation the network protocol used in Japanese scientific satellites had the overlaps in each protocol layer skillfully removed. This indicates that the adjustment for each protocol layer was done adeptly under close communication among the parties involved when the specications were compiled. We checked that RMAP itself was a protocol with sufficient function in providing real-time operation capability. Therefore, we pointed out that if we utilized this data format and communication protocol, SpaceWire-RT or SpaceWire-T were unnecessary in maintaining the real-time property, and imprinted our existing Japanese development specs in the form of an improvement proposal.[17] Moreover, Small Demonstration Satellite 1 (SDS-1) was launched in 2009, and the specification that we proposed was successfully demonstrated in orbit.Figure 2 shows the communication standard layer that we proposed. We showed that what was initially done in eight layers or more could be done in seven layers, as shown in Fig. 2. The advantage of this communication standard layer is that it provides the real-time functionality required in onboard network for satellites by a simple protocol, based on the experience of development and operation of scientic satellites. The draft proposal was submitted at the 15th SpW WG in 2010 and obtained unanimous approval from the participating countries of ESA/ESTEC. This realized the simplification of the SpaceWire communication standard layers, and scalability could be realized from small satellites of 100 kg level to large satellites of 2.7 tons. We believe the simple and high-performance characteristic of SpaceWire could not have been obtained if this proposal was not submitted from Japan.Another advantage is the point that we were able to respect the positions of the participants of all countries even in midst of the standardization proposal. As mentioned before, we swiftly launched SDS-1 in 2009 and succeeded in demonstrating the SpaceWire RMAP standard in orbit for the first time in the world. At this point, we felt for the first time that ESA trusted us. However, it was not because our technological level was demonstrated. They worried whether the standard on which they were working would be operational in orbit, or not. Instead of reporting that Japan was successful in onboard demonstration of the SpaceWire RMAP in orbit, we reported that the draft standard specification on which we were working that was the result of SpW WG was successfully demonstrated in orbit. In consequence, the results of the orbital demonstration by Japan wiped out their worries, and its success was shared by all parties involved. It seemed this led to the trusting relationship.Fig. 2 Communication standard layer proposed to SpW WG from Japan[17]User ApplicationPlug and PlayPacket Transfer ProtocolSegmentation/BlockingRetry/RedundancyProtocol ID / RMAP(ECSS-E-ST-50-51C/52C)Scheduling(SpaceWire-D Draft B)SpaceWire(ECSS-E-ST-50-12C)

元のページ  ../index.html#41