Technologies and methodologies for CubeSat performances improvement

The importance of small satellites, and in particular of nano-satellites (e.g. CubeSats) has increased during last years thanks to major improvements in the field of electrical and mechanical miniaturisation. Another important factor is represented by the interest of national and international space agencies, which has led to the creation of many scientific small satellites programmes, beyond several educational ones. “Small satellites” term shall not be considered referring only to mass, but it describes also a new approach to building, operating, and managing risk for satellite systems; in fact, CubeSat standard (whose reference design was proposed in 1999, and first launch occurred in 2003) is becoming a concrete realisation of a new way of thinking space systems, which is changing the way to access the space. Commercial components, rapid scheduling, risk tolerance and lean testing are just some of features behind the success and spread of CubeSats among universities, space agencies, research and scientific centres, and private companies. Current CubeSat and small satellite missions are mostly developed for Low Earth Orbit (LEO) application, and the number of scientific goals/tasks that they can perform is still limited. CubeSats are nowadays a mature technology to perform Earth observation with low-to-medium performance, and they are a valid educational tool to train young engineers and students in the process of conceiving, implementing and operating a space mission. However, it is possible to say we are entering a new CubeSat Era, in which CubeSats will be called to carry out real missions of the future. Within the new framework, they may represent new test-beds for future bigger missions or allow independent and unprecedented new applications. This research aims at contributing to advance the state of the art in CubeSat missions design and implementation by enhancing some technologies that will support those missions and by defining some innovative approaches for CubeSat development. This main objective has been addressed and pursued from two different points of view: -design perspective, to contribute at improving one specific technology in one technical domain of interest at sub-system level. A subsystem, specifically the Attitude Determination and Control Subsystem (ADCS), has been chosen and the attitude determination process is the function that has been specifically analysed -development perspective, to contribute at improving the development process of a CubeSat mission. This activity has been carried out at system-level, and addresses specifically the Assembly, Integration and Test/Verification (AIT/V) process of a CubeSat for which a new approach has been proposed based on lessons learned from past missions and innovative simulation methodologies and tools. The problem definition for this thesis has been expressed with the following questions: “How and to what extent can the CubeSat platform support future space missions for science purposes, technology demonstration, and service applications?”, and “What features of CubeSat platforms and their missions shall be improved to meet the emerging needs and requirements?”. To answer these questions, the whole CubeSat life-cycle has been considered, analysed and eventually adapted to the rising needs. Both design aspects and development processes have been addressed, which might help improving overall CubeSat quality, extend CubeSat applications range, and finally increase mission success. The major results are represented by improvements attainable at different phases of the CubeSat life-cycle, such as design, development and verification phase, and at different levels (i.e. subsystem level and system level), through the use of In-the-Loop (IL) simulator and leveraging lessons learned and heritage from previous missions. The methodology adopted is the Model Based System Engineering (MBSE), which provides a wonderful support to solve complex problems against reduced budgets, fewer resources (in terms of personnel and money) and shorter schedules. The first part of the work has been aimed at the investigation of the CubeSat standard and its diffusion, in order to identify actions to improve performances in view of next commercial and scientific missions. Furthermore, a detailed analysis of state-of-the-art and on-the-horizon technologies has been carried out, with a major focus on ADCS, COMmunication subSYStem (COMSYS), Electrical Power Subsystem (EPS) and propulsion subsystem. For what concerns the subsystem level, the ADCS has been selected as case study, with a direct application on e-st@r-II CubeSat. Determination algorithms have been investigated, both static and recursive ones, and a specific recursive algorithm have been designed, developed and integrated on-board. The algorithm works in the special condition of under-observability: a single vector observation is available, which is the Earth Magnetic Field (EMF) vector. The algorithm is part of the payload of e-st@r-II CubeSat, which was selected by ESA Education Office to participate to Fly Your Satellite! (FYS!) initiative and whose launch is scheduled on 22nd April 2016 on board Soyuz ST-A VS14 from Centre Spatial Guyanais (CSG) in French Guyana. The on-orbit testing will provide additional data to validate the algorithm. In addition to this activity, the adoption of Artificial Neural Network (ANN) technology on-board CubeSats has been investigated. The field of application of ANNs is again the ADCS: they have been designed to act as state estimator and fault detector. Several types of ANNs have been studied, in order to identify the ones with the best performances. In this application, pattern recognition neural networks and Many ADAptive LInear NEuron (MADALINE) networks have been implemented to detect and identify a fault of a gyro and to estimate angular velocities and attitude of the satellite even when a fault occurs. For what concerns the system level, the goal of performances improvement has been pursued working not only on the design phase, but also improving the other phases, such as development, manufacturing, assembly, integration, verification and operations phases. Key aspects have been identified regarding the verification campaign for CubeSats: reference standards can not be adopted integrally, but there is a need for tailoring, creating a set of lean tests, and the use of IL simulators, in all of their forms (Algorithm-In-the-Loop (AIL), Software-In-the-Loop (SIL), Controller-In-the-Loop (CIL) and Hardware-In-the-Loop (HIL)), is fundamental, especially for verifications that could be very demanding for the flight hardware and/or very expensive. These aspects have been discussed, with reference to e-st@r-II CubeSat functional and environmental test campaigns within the participation at FYS! programme of ESA Education Office. Finally, an additional possible way to increase the rate of success has been found into start thinking as a “manufacturing company” (i.e. in terms of mass production), and to evaluate previous missions, gathering lessons learned, extracting possible failures and drawbacks, and deducing possible improvements for future projects, at all phases of the product life-cycle. As example of this methodology, the improvements achieved on e-st@r-II CubeSat thanks to e-st@r-I CubeSat have been presented, together with the lessons learned of the environmental test campaign of e-st@r-II CubeSat performed at ESA-ESTEC, which will have an impact on next projects of the Team (e.g. 3-STAR CubeSat). In conclusion, it has been proven that the proposed technologies and methodologies methods are effective to increase the CubeSats mission success and their performances, investigating some direct applications. Moreover, in view of future missions of CubeSats, some recommendations have been formulated, and they may be useful both for educational programs and scientific/commercial ones.