Hierarchical control of intelligent machines applied to spacestation telerobots

A hierarchical architecture is described which supports space station telerobots in a variety of modes. The system is divided into three hierarchies: task decomposition, world model, and sensory processing. Goals at each level of the task decomposition hierarchy are divided both spatially and temporarily into simpler commands for the next lower level. This decomposition is repeated until, at the lowest level, the drive signals to the robot actuators are generated. To accomplish its goals, task decomposition modules must often use information stored in the world model. The purpose of the sensory system is to update the world model as rapidly as possible to keep the model in registration with the physical world. The architecture of the entire control system hierarchy and how it can be applied to space telerobot applications is described     

Lessons learned from the dynamic identification/qualification tests on the ESC-A upper stage model

The dynamic qualification of the new cryogenic upper stage ESC-A of the ARIANE 5 is supported by several tests in order to verify the assumptions and the modeling approach made at the beginning of the development. The stage contains a large amount of equipment such as propellant lines, acceleration rockets, batteries, fluid control equipment etc. For the low frequency domain the verification of the equipment responses in the integrated state was done by a sine vibration test, excited to levels representing the predicted flight loads including a qualification factor. Acoustic tests with the upper stage were performed to verify the random vibration responses in the frequency range up to 2000 Hz. To verify the shock response level induced by stage separation (pyro-shock) a stage separation test was performed. The paper concentrates on the experience made with the modal identification and sine vibration test of the stage. For the sine vibration test an electro-dynamic multi-shaker table was used. It was able to produce the required input precisely up to as specified, not an easy task for a test set-up of 20 tons weight. The paper presents the approach of how the dynamic qualification was reached successfully and highlights the experience accomplished.