Session
Technical Session III: Advanced Sensor Concepts
Abstract
In June 2000, the Surrey Space Centre (SSC) and Surrey Satellite Technology Limited (SSTL) launched the remote inspection demonstrator nanosatellite, "SNAP-1". One of the primary mission objectives of this satellite was to image its companion microsatellite, Tsinghua-1, during the deployment phase of the launch. Later in the mission it is also planned that SNAP-1 will be manoeuvred back within visual range of Tsinghua-1, in order to carry out further imaging experiments whilst the satellites fly in formation. To fulfill its mission, SNAP-1 carries a powerful, innovative and highly integrated Machine Vision System (MVS). This consists of four ulta-minature CMOS video cameras, a "software" video digitiser, 8Mb of 70ns SRAM and a 220MHz StrongARM processor. The integration of these components provides a low cost, low power consumption, high reliability platform, with enough processing power to capture and process real-time video images. This will enable SNAP-1 to not only compress and return images back to Earth, but to perform high level computer vision functions such as optical target tracking, automatic pose and position estimation and on future SNAP missions perhaps even optically guided docking. This paper therefore details the design, performance and initial results from the SNAP-1 Machine Vision System.
The SNAP-1 Machine Vision System
In June 2000, the Surrey Space Centre (SSC) and Surrey Satellite Technology Limited (SSTL) launched the remote inspection demonstrator nanosatellite, "SNAP-1". One of the primary mission objectives of this satellite was to image its companion microsatellite, Tsinghua-1, during the deployment phase of the launch. Later in the mission it is also planned that SNAP-1 will be manoeuvred back within visual range of Tsinghua-1, in order to carry out further imaging experiments whilst the satellites fly in formation. To fulfill its mission, SNAP-1 carries a powerful, innovative and highly integrated Machine Vision System (MVS). This consists of four ulta-minature CMOS video cameras, a "software" video digitiser, 8Mb of 70ns SRAM and a 220MHz StrongARM processor. The integration of these components provides a low cost, low power consumption, high reliability platform, with enough processing power to capture and process real-time video images. This will enable SNAP-1 to not only compress and return images back to Earth, but to perform high level computer vision functions such as optical target tracking, automatic pose and position estimation and on future SNAP missions perhaps even optically guided docking. This paper therefore details the design, performance and initial results from the SNAP-1 Machine Vision System.
