On June 2nd 2003 the Beagle 2
lander was launched aboard a Soyuz-Fregat rocket, and was planned to arrive on Mars on December 25th 2003. The Beagle 2 program was a British led quest to land on Mars as a part of ESA's
Mars Express mission.
The primary aim of Beagle 2 was to search for evidence of past life on Mars. A key component in this science task was a robot arm which was equipped with an end effector containing a set of instruments to examine the Martian surface, sub-surface, and nearby rocks.
The robot ARM together with its end effector, called the PAW, must be positioned correctly so that the instruments can function properly, and part of the process required to achieve the necessary position accuracy and repeatability involved calibrating the robot ARM prior to launch. The Beagle 2 lander, ARM and PAW were designed with the aid of the Catia
CAD software suite.
Mission operations required a software simulation of the lander and robot arm so that instrument deployment, and other mission tasks, could be exhaustively studied prior to engagement of the real Beagle 2 hardware on Mars. UWA were members of the Beagle 2 consortium,
and calibrated the Beagle 2 ARM, and built the software simulation of the lander, ARM and PAW, for mission validation and operation purposes.
A short AVI movie of an early Beagle 2 simulation is available (7MB zipped).
Our involvement represented a unique opportunity to participate in an historic Mars mission. Funding for the calibration and simulation work was from the 1999 HEFCW Research Infrastructure Initiative, the 2001/2002 JREI Enhancement of Capital Grant for Research Infrastructure Allocation, EPSRC Studentship Awards, and UWA.
The traditional approach to space vehicle operations is for humans to carry out on the ground, a large number of functions. These include planning, sequencing space vehicle actions, tracking internal hardware state, ensuring correct functioning, recovering in cases of failure, and subsequently working around faulty subsystems. This approach will no longer be viable in the future due to a) increased round trip light time for ‘hands-on’ operations to be effective, and b) a desire to limit the operations team and space communications costs. In short, future robots operating in remote harmful environments such as space, will require greater autonomy. However, this ability will not only apply to the control of the robot, but it will require also that such devices are capable of detecting internal error states, diagnosing their causal faults, and where possible, recovering from such conditions, all without human intervention. Clearly, this is a tall order for all possible situations, and one must focus upon problem sub-sets if knowledge advances are to be made. We are addressing the problem area of robot arm fault diagnosis and recovery, and are focusing upon the area of mechanical damage, its detection, diagnosis, and where possible, its recovery. We seek to deliver results that are of generic value to the science areas of qualitative kinematics and model based diagnosis for robots operating in remote harmful environments. This research is being funded by the UK Engineering and Physical Sciences Research Council. EPSRC Grant: GR/S10766/01 - "Model based methods for kinematics fault diagnosis and remediation".
The information provided on this and other pages by me, Dave Barnes, is under my own personal responsibility and not that of Aberystwyth University. Similarly, any opinions expressed are my own and are in no way to be taken as those of Aberystwyth University.