Plan Execution for Autonomous Spacecraft

In this paper, we propose a stratified agent architecture for an autonomous earth imaging spacecraft to achieve higher levels of autonomy. The autonomous spacecraft should possess the self-adaptation and the self-healing capabilities, in addition to other fundamental autonomy requirements including reflex, awareness and self-regulation. The architecture is configured as stratified agent architecture with two strata. The upper stratum is concerned with the adaptive behaviour, which is achieved through an on-board mission planner. This stratum is in turn configured as hierarchical agent architecture. Hierarchical planning in a Petri net framework involves refinement of a transition in the upper layer using a Petri net model and searching for a valid plan using the reachability analysis technique. The lower stratum deals with the execution aspects and provides the reactive behaviour. It is configured with three peer level agents to achieve the reflex and the awareness properties. These agents follow a generic structure with perception, action, and communication components. The agent achieves the required autonomy capabilities by using an appropriate knowledge model. The uniqueness of the architecture is the utilisation of Petri net models, with varying degree of abstraction, in the various layers to represent the planning knowledge, diagnostic knowledge and the appropriate world model. In this paper, we present how the various capabilities for autonomy are achieved by the agent organisation by exploiting the Petri net models.

Proceedings of the second international conference on Autonomous agents - AGENTS '98, 1998

The New Millennium Remote Agent (NMRA) will be the first AI system to control an actual spacecraft. The spacecraft domain places a strong premium on autonomy and requires dynamic recoveries and robust concurrent execution, all in the presence of tight real-time deadlines, changing goals, scarce resource constraints, and a wide variety of possible failures. To achieve this level of execution robustness, we have integrated a procedural executive based on generic procedures with a deductive model-based executive. A procedural executive provides sophisticated control constructs such as loops, parallel activity, locks, and synchronization which are used for robust schedule execution, hierarchical task decomposition, and routine configuration management. A deductive executive provides algorithms for sophisticated state inference and optimal failure recovery planning. The integrated executive enables designers to code knowledge via a combination of procedures and declarative models, yielding a rich modeling capability suitable to the challenges of real spacecraft control. The interface between the two executives ensures both that recovery sequences are smoothly merged into high-level schedule execution and that a high degree of reactivity is retained to effectively handle additional failures during recovery.

Proceedings of the Fifth …, 2000

On May 17th 1999, NASA activated for the first time an AI-based planner/scheduler running on the flight processor of a spacecraft. This was part of the Remote Agent Experiment (RAX), a demonstration of closed-loop planning and execution, and model-based state in-ference and failure ...

Space Sciencecraft Control and Tracking in the New Millennium, 1996

NASA has recently announced the New Millennium Program NMP to develop faster, better, cheaper" spacecraft in order to establish a virtual presence" in space. A crucial element i n a c hieving this vision is onboard spacecraft autonomy, requiring us to automate functions which h a v e traditionally been achieved on ground by humans. These include planning activities, sequencing spacecraft actions, tracking spacecraft state, ensuring correct functioning, recovering in cases of failure and recon guring hardware. In response to these challenging requirements, we analyzed the spacecraft domain to determine its unique properties and developed an architecture which provided the required functionality. This architecture integrates traditional real-time monitoring and control with constraint-based planning and scheduling, robust multi-threaded execution, and model-based diagnosis and recon guration. In a ve month e ort we successfully demonstrated this implemented architecture in the context of an autonomous insertion of a simulated spacecraft into orbit around Saturn, trading o science and engineering goals, and achieving the mission goals in the face of any single point of hardware failure. This scenario turned out to be among the most complex handled by each of the component technologies. As a result of this success, the integrated architecture has been selected to control the rst NMP ight, Deep Space One, in 1998. It will be the rst AI system to autonomously control an actual spacecraft.

1996

NASA has recently announced the New Millennium Program (NMP) to develop \faster, better, cheaper" spacecraft in order to establish a \virtual presence" in space. A crucial element in achieving this vision is onboard spacecraft autonomy, requiring us to automate functions which have traditionally been achieved on ground by humans. These include planning activities, sequencing spacecraft actions, tracking spacecraft state, ensuring correct functioning, recovering in cases of failure and recon guring hardware. In response to these challenging requirements, we analyzed the spacecraft domain to determine its unique properties and developed an architecture which provided the required functionality. This architecture integrates traditional real-time monitoring and control with constraint-based planning and scheduling, robust multi-threaded execution, and model-based diagnosis and recon guration. In a ve month e ort we successfully demonstrated this implemented architecture in the context of an autonomous insertion of a simulated spacecraft into orbit around Saturn, trading o science and engineering goals, and achieving the mission goals in the face of any single point of hardware failure. This scenario turned out to be among the most complex handled by each of the component technologies. As a result of this success, the integrated architecture has been selected to control the rst NMP ight, Deep Space One, in 1998. It will be the rst AI system to autonomously control an actual spacecraft.

The Deep Space One (DS1) mission, scheduled to fly in 1998, will be the first spacecraft to feature an on-board planner. The planner is part of an artificial intelligence based control architecture that comprises a planner/scheduler, a plan execution engine, and a model-based fault diagnosis and reconfiguration engine. This autonomy architecture reduces mission costs and increases mission quality by enabling highlevel commanding, robust fault responses, and opportunistic responses to serendipitous events. This paper describes the on-board planning and scheduling component of the DS1 autonomy architecture.

2000

This paper describes the ASPEN system for automation of planning and scheduling for space mission operations. ASPEN contains a number of innovations including: an expressive but easy to use modeling language, multiple search (inference) engines, iterative repair suited for mixed-initiative human in loop operations, real-time replanning and response (in the CASPER system), and plan optimization. ASPEN is being used for the Citizen Explorer (CX-1) (August 2000 launch) and the 2 nd Antarctic Mapping Missions (AMM-2) (September 2000). ASPEN has also been used to automate ground communications stations -automating generation of tracking plans for the Deep Space Terminal (DS-T). ASPEN has been used to demonstrate automated command generation and onboard planning for rovers and is currently being evaluated for operational use for the Mars-01 Marie Curie rover mission. CASPER, the soft real-time versions of ASPEN, has been demonstrated with the Jet Propulsion Laboratory (JPL) Mission Data Systems (MDS) Control Architecture prototypes. .

1998

This paper describes the Remote Agent flight experiment for spacecraft commanding and control. In the Remote Agent approach, the operational rules and constraints are encoded in the flight software. The software may be considered to be an autonomous “remote agent” of the spacecraft operators in the sense that the operators rely on the agent to achieve particular goals. The experiment will be executed during the flight of NASA's Deep Space One technology validation mission. During the experiment, the spacecraft will not be given the usual detailed sequence of commands to execute. Instead, the spacecraft will be given a list of goals to achieve during the experiment. In flight, the Remote Agent flight software will generate a plan to accomplish the goals and then execute the plan in a robust manner while keeping track of how well the plan is being accomplished. During plan execution, the Remote Agent stays on the lookout for any hardware faults that might require recovery actions or replanning. In addition to describing the design of the remote agent, this paper discusses technology-insertion challenges and the approach used in the Remote Agent approach to address these challenges. The experiment integrates several spacecraft autonomy technologies developed at NASA Ames and the Jet Propulsion Laboratory: on-board planning, a robust multi threaded executive, and model-based failure diagnosis and recovery