ESA GNC Conference Papers Repository

The ’intelligent attitude and orientation control system’, iAOCSTM, is a bottom-up approach to the problem of designing and implementing an modular AOCS for small satellites. Usually the development and design of a satellite AOCS is mission driven, i.e. the AOCS is set up as an individual system and adjusted with great effort to the mission respectively the satellite (platform). Once the requirements are met, the sensors and actuators are selected and positioned to meet the derived requirements, as well as the mass, size and cost budget. Todays AOCS are highly specific to a given satellite and/or mission if not even mission phase and/or AOCS mode during that phase. As a result, re-use of software from mission to mission and sometimes even between spacecraft, where (slightly) different sets of sensor or actuators are used, is almost impossible. Furthermore, on-board systems of spacecraft continue to increase in complexity, have to meet stricter requirements and must provide more functionality. At the same time there is a strong need to reduce the (re-)design effort, the validation complexity and increase the test reliability, not only for the AOCS software itself, but also for all areas related to fault detection, isolation and recovery (FDIR) from individual components to the system level. It is clear that a highly flexible ’reference architecture’ could help with many of these issues if it is based on a small number of modular building blocks, where each module can be developed, configured, tested, qualified and used by itself as well as together with others. Unfortunately, the current approach to AOCS is anything but modular or flexible or simple, which is in part due to a lack of standardisation of the interface or communication between components, but mostly due to the fact that traditional AOCS design requires a tight coupling between the individual components and the central AOCS core. For this reason we have investigated the possibility of designing an AOCS system from the bottom up, where system and component reliability, availability, safety and COTS based tests are not ’bolted onto’ an existing design, but where these considerations are the design drivers for the software and hardware architecture; and where testing and calibration are not the last steps in the development process, but where automatic testing and procedures for self recalibration are part of the development process from the start. In our paper we will present the key parts of an intelligent AOCS (iAOCS), which promises to overcome these limitations. The iAOCS consists of smart sensors, an intermediate ’data fusion’ layer, a simplified AOCS core and smart actuators.