No abstract

Vision-based navigation has become increasingly important in a variety of space applications for enhancing autonomy and dependability. Future missions, such as active debris removal for remediating the low Earth orbit environment, will rely on novel high-performance avionics to support advanced image processing algorithms with substantial workloads. However, when designing new avionics architectures, constraints relating to the use of electronics in space present great challenges, further exacerbated by the need for significantly faster processing compared to conventional space-grade central processing units. With the long-term goal of designing highperformance embedded computers for space, in this paper, an extended study and tradeoff analysis of a diverse set of computing platforms and architectures (i.e., central processing units, multicore digital signal processors, graphics processing units, and field-programmable gate arrays) are performed with radiation-hardened and commercial offthe-shelf technology. Overall, the study involves more than 30 devices and 10 benchmarks, which are selected after exploring the algorithms and specifications required for vision-based navigation. The present analysis combines literature survey and in-house development/testing to derive a sizable consistent picture of all possible solutions. Among others, the results show that certain 28 nm system-on-chip devices perform faster than space-grade and embedded central processing units by 1-3 orders of magnitude, while consuming less than 10 W. Field-programmable gate array platforms provide the highest performance per watt ratio.

In recent years, research in the space community has shown a growing interest in Artificial Intelligence (AI), mostly driven by systems miniaturization and commercial competition. In particular, the application of Deep Learning (DL) techniques on board Earth Observation (EO) satellites might lead to numerous advantages in terms of mitigation of downlink bandwidth constraints, costs, and increment of the satellite autonomy. In this framework, the CloudScout project, funded by the European Space Agency (ESA), represents the first time in-orbit demonstration of a Convolutional Neural Network (CNN) applied to hyperspectral images for cloud detection. The first instance of this use case has been done with an INTEL Myriad 2 VPU on board a CubeSat optimized for low cost, size, and power efficiency. Nevertheless, this solution introduces multiple drawbacks due to its design not specifically being for the space environment, thus limiting its applicability to short-lifetime Low Earth Orbit (LEO) applications. The current work provides a benchmark between the Myriad 2 and our custom hardware accelerator designed for Field Programmable Gate Arrays (FPGAs). The metrics used for comparison include inference time, power consumption, space qualification, and components. The obtained results show that the FPGA-based solution is characterized by a reduced inference time, and a higher possibility of customization, but at the cost of greater power consumption and a longer Time to Market. As a conclusion, the proposed approach might extend the potential market of DL-based solutions to long-term LEO or interplanetary exploration missions through deployment on space-qualified FPGAs, with a limited cost in energy efficiency.

Artificial intelligence (AI) is paving the way for a new era of algorithms focusing directly on the information contained in the data, autonomously extracting relevant features for a given application. While the initial paradigm was to have these applications run by a server hosted processor, recent advances in microelectronics provide hardware accelerators with an efficient ratio between computation and energy consumption, enabling the implementation of AI algorithms "at the edge." In this way only the meaningful and useful data are transmitted to the end-user, minimizing the required data bandwidth, and reducing the latency with respect to the cloud computing model. In recent years, European Space Agency (ESA) is promoting the development of disruptive innovative technologies on-board earth observation (EO) missions. In this field, the most advanced experiment to date is the -sat-1, which has demonstrated the potential of artificial intelligence (AI) as a reliable and accurate tool for cloud detection on-board a hyperspectral imaging mission. The activities involved included demonstrating the robustness of the Intel Movidius Myriad 2 hardware accelerator against ionizing radiation, developing a Cloudscout segmentation neural network (NN), run on Myriad 2, to identify, classify, and eventually discard on-board the cloudy images, and assessing the innovative Hyperscout-2 hyperspectral sensor. This mission represents the first official attempt to successfully run an AI deep convolutional NN (CNN) directly inferencing on a dedicated accelerator on-board a satellite, opening the way for a new era of discovery and commercial applications driven by the deployment of on-board AI.

Traditional heavy-ion testing for single-event effects is carried out in cyclotron facilities with energies around 10 MeV/n. Despite their capability of providing a broad range of linear energy transfer (LET) values, the main limitations are related to the need of testing in a vacuum and with the sensitive region of the components accessible to the low range ions. In this paper, we explore the use of ultrahigh energy (UHE) (5-150 GeV/n) ions in the CERN accelerator complex for radiation effects on electronics testing. At these energies, we show, both through simulations and experimental data, the significant impact of the ion energy on the ionization track structure and associated volume-restricted LET value, highlighting the possible limitations for radiation hardness assurance for highenergy accelerator applications. In addition, we show that from a nuclear interaction perspective, UHE ions behave similar to protons independently of their significantly larger mass.