Aerospace work at AIAI

[A page from 1997]

AIAI has made significant contributions to the application of artificial intelligence in the aerospace industry.

AIAI has worked on a number of projects for the European Space Agency (ESA). These have included:

• Joint work with partners CRI (Denmark) and Matra Espace (France) on the PlanERS system which generates operational sequences for the European ERS-1 spacecraft. The work also studied the use of AI planning and scheduling technologies (in which AIAI is a world leader) for a wide range of other ESA mission opportunities both manned and unmanned.
• Another system, Optimum-AIV, was developed by a consortium involving CRI, Matra Espace, ProgeSpace(UK) and AIAI. This knowledge based planning system assisted in planning and scheduling activities for spacecraft assembly, integration and verification for a system such as ERS-1. The system is now in active use for planning the assembly and test of the payload bay of Ariane rockets.
• AIAI's Technical Director, Professor Austin Tate, was a member of an ESA expert panel which advises on telematics and automation technologies and involved in the original studies on the PROBA spacecraft.

AIAI also worked on early use of KBS methods with the Darmstadt space operations centre (ESOC).

AIAI has worked with British Aerospace on the EUMETSAT contract to build a mission planning and scheduling module for the European Organisation for the Exploitation of Meteorological Satellites. This developed the spacecraft telecommand and scheduling systems for the overall control centre.

The UK Defence Research Agency-Satellite Communication Division commissioned AIAI to undertake a feasibility study on whether an AI based planning tool could be developed to improve the network planning and access planning capability for UK military satellite communications. The ultimate objective was for a tool to assist operators manage and plan the satellite communications network more effectively, making optimal use of available resources.

As far as our work in the United States is concerned, AIAI's O-Plan system forms a key part of an ambitious long term research and development programme with the aim of producing a flexible support tool for command, planning and control in applications such as project management, logistics and spacecraft operations. Early work was for the USAF Rome Laboratory and involved spacecraft mission planning.

Early work in the 1970s by AIAI on the Nonlin planner was utilised in experimental planning systems for the Voyager spacecraft at NASA JPL (the resulting system was called DEVISER). JPL are now actively using advanced planning technologies from AI in the Deep Space One autonomous spacecraft project.

AIAI was part of a 7 partner UK consortium led by the British National Space Centre's academic arm at the Rutherford Lab Space Division to investigate a technology proving satellite called T-SAT. AIAI investigated on-board autonomy for control, spacecraft health monitoring and for "quick look" on-board image processing. In collaboration with the University of Surrey (who were part of the T-SAT team), the project also investigated ground segment mission planning. This led to a demonstrator schedulng system caled T-SCHED which actually generated on-board computer diary commands for the UoSAT-II spacecraft and its instruments.