摘要 :
As autonomous systems become more prevalent, it is crucial to develop new methods for ensuring their safety. The National Aeronautics and Space Administration (NASA)'s Robust Software Engineering (RSE) group is addressing this nee...
展开
As autonomous systems become more prevalent, it is crucial to develop new methods for ensuring their safety. The National Aeronautics and Space Administration (NASA)'s Robust Software Engineering (RSE) group is addressing this need with the development of the Research for Autonomous Vehicles (R-RAV) project, an autonomous rover testbed designed for assured autonomy research. In this paper, we describe how we used a Model-Based Systems Engineering (MBSE) approach to design and build the R-RAV and implemented our first autonomy research mission. The adoption of MBSE has allowed for efficient and data-driven collaboration, and has provided a comprehensive view of the system throughout its development, reducing ambiguity while increasing traceability and productivity. The R-RAV testbed will be used to advance research in assured autonomy, including safe machine learning, automated testing, and formal verification.
收起
摘要 :
As autonomous systems become more prevalent, it is crucial to develop new methods for ensuring their safety. The National Aeronautics and Space Administration (NASA)'s Robust Software Engineering (RSE) group is addressing this nee...
展开
As autonomous systems become more prevalent, it is crucial to develop new methods for ensuring their safety. The National Aeronautics and Space Administration (NASA)'s Robust Software Engineering (RSE) group is addressing this need with the development of the Research for Autonomous Vehicles (R-RAV) project, an autonomous rover testbed designed for assured autonomy research. In this paper, we describe how we used a Model-Based Systems Engineering (MBSE) approach to design and build the R-RAV and implemented our first autonomy research mission. The adoption of MBSE has allowed for efficient and data-driven collaboration, and has provided a comprehensive view of the system throughout its development, reducing ambiguity while increasing traceability and productivity. The R-RAV testbed will be used to advance research in assured autonomy, including safe machine learning, automated testing, and formal verification.
收起
摘要 :
As autonomous systems become more prevalent, it is crucial to develop new methods for ensuring their safety. The National Aeronautics and Space Administration (NASA)'s Robust Software Engineering (RSE) group is addressing this nee...
展开
As autonomous systems become more prevalent, it is crucial to develop new methods for ensuring their safety. The National Aeronautics and Space Administration (NASA)'s Robust Software Engineering (RSE) group is addressing this need with the development of the Research for Autonomous Vehicles (R-RAV) project, an autonomous rover testbed designed for assured autonomy research. In this paper, we describe how we used a Model-Based Systems Engineering (MBSE) approach to design and build the R-RAV and implemented our first autonomy research mission. The adoption of MBSE has allowed for efficient and data-driven collaboration, and has provided a comprehensive view of the system throughout its development, reducing ambiguity while increasing traceability and productivity. The R-RAV testbed will be used to advance research in assured autonomy, including safe machine learning, automated testing, and formal verification.
收起
摘要 :
The next-generation Airborne Collision Avoidance System for smaller UASs (ACAS sXu) is currently being developed and tested by the Federal Aviation Administration (FAA) to provide detect-and-avoid capability for small unmanned air...
展开
The next-generation Airborne Collision Avoidance System for smaller UASs (ACAS sXu) is currently being developed and tested by the Federal Aviation Administration (FAA) to provide detect-and-avoid capability for small unmanned aircraft operating beyond line-of-sight. Due to the complexity and safety-critical nature of the system, safety validation is important not only for the certification of the final system, but also for informing changes during the iterative development process. In this paper, we analyze a prototype of ACAS sXu in simulated aircraft encounters to discover scenarios of small near mid-air collisions (sNMACs), an important safety event in which two aircraft come closer than 50 feet horizontally and 15 feet vertically. Due to the size and complexity of the system as well as rarity of sNMAC events, traditional methods such as Monte Carlo testing often require informed setup and targeting to elicit failures. However, such a dependence on domain knowledge can be incompatible with the independent verification and validation (IV&V) process, the aim of which is to discover unforeseen issues. To address these challenges, we apply an accelerated validation method called adaptive stress testing (AST) to find the most likely sNMAC scenarios without reliance on system introspection. AST uses reinforcement learning to adapt the search towards the most promising areas of the search space as it progresses. We use a state-of-the-art deep reinforcement learning algorithm, proximate policy optimization, to more efficiently search the large and continuous state space. We find that this approach significantly improves the performance of AST compared to a prior approach based on Monte Carlo tree search. We perform experiments using AST to find sNMAC events under various encounter configurations, varying parameters pertaining to dynamics and coordination. Our experiments show AST to be very effective at finding sNMAC scenarios. We summarize our findings, presenting high-level categories of discovered sNMACs and specific examples of encounters in each category.
收起
摘要 :
Recently, a lot of attention has been brought to the deep space mining missions. Mining of the natural resources from near-Earth celestial bodies like Moon, near-Earth asteroids (NEAs), comets and meteors is one of the possible wa...
展开
Recently, a lot of attention has been brought to the deep space mining missions. Mining of the natural resources from near-Earth celestial bodies like Moon, near-Earth asteroids (NEAs), comets and meteors is one of the possible ways for solving of the problem of lack of natural resources on Earth. The calculations in the article show that current cost of the launch and space mining technology readiness level do not allow us to implement extraterrestrial resources to Earth economy, due to the extremely high cost comparing to the terrestrial resources. The goal of this paper is to estimate economical benefits of asteroid mining for Moon and Mars colonies. We know from planetary research programs that water resources on the Moon are limited and most of them are possibly concentrated near the South Pole. On Mars, we know that water ice is concentrated near the Polar Regions in form of ice (Polar Layered Deposits) or near-surface permafrost. In order to enable in-situ resource utilization (fuel generation, building materials etc) a number of additional materials have to be brought to the Moon or Mars. We propose to consider a network economy that includes Near-Earth Asteroids (NEA), Earth, Moon, asteroid belt, Mars and its satellites as a source of materials for Solar System colonization. For example, materials contained in NEA may be very useful to this task. Approximately 10% of known NEAs are more accessible than the Moon in terms of required delta-V for transfer. Moreover, asteroids are a possible source of platinum group metals (PGM) and water according to the ground observations. To simulate scenarios on colony growth, material transfer and asteroid exploitation, we would like to utilize multi-agent network optimization. We can map agents acting in the Solar System economy to well proven solutions in dynamic optimization for truck delivery systems. We have constructed a simple ontology that includes colonies (resource requests), spacecraft (resource transport), ast
收起