Many mechanical tasks in the microgravity environment on space vehicles and space stations can be efficiently carried out by robotic manipulators. Not only can the objective of “minimal intervention by crew members” be satisfied in this manner, it is also possible to meet various task specifications in the dynamic environment of a space application more effectively (particularly in terms of time, precision, and reliability) by employing robots.
Tasks of interest include delicate experiments as well as production and maintenance operations in space. High load-capacity/mass ratio, autonomous operation, high accuracy and repeatability, high stiffness, and high dexterity are some of the generally preferred characteristics for robotic manipulators used in space applications. Gear transmission at joints is known to introduce undesirable backlash resulting in low stiffness, degraded accuracy and repeatability and high friction with associated high levels of power dissipation and thermal and wear problems. Direct-drive manipulation appears to reduce these problems, but in this case, manipulator joints tend to be rather massive. The traction-drive principle developed by NASA promises improvements in this direction, while providing gearless transmission.
The base reactions of a space manipulator are directly transmitted to the supporting structure, which is generally
a part of the space vehicle or space station. These dynamic forces (and torques) are in fact disturbances on the supporting structure, as well as on other equipment and operations in the robot’s environment. Furthermore, since the base reactions represent dynamic coupling between the robot and the space structure, not only will the environment be affected by these disturbances, but also the performance of the robot itself. It is not trivial to take into account this coupling in the control schemes for a space structure and for a space robot. Ideally, one would desire zero base reactions, but in practice, minimization of an appropriate cost function would be acceptable. This latter approach has been taken by us. Specifically, the redundant degrees of freedom in a redundant robot were employed to dynamically minimize a quadratic cost function of base reactions. A four-degree-of-freedom robot having two traction-drive joints has been studied by us using this approach, providing encouraging results. Another aspect that requires attention is the handling of disturbance sensitive specimens in space. The approach taken by us was to design the end effector trajectory of a robotic task in space such that acceleration and jerk are constrained while meeting the desired time and position objectives of the particular task. Specifically, cycloidal trajectories were employed.
There are several research and project-specific issues that have to be addressed under space-station robotics. Some are dynamic analysis and design issues pertaining to space robotics and some others are associated control issues. Several of these issues are as follows:
1. Effects of unplanned influences, payload variations and disturbances (e.g., obstacles and collisions) on operating conditions and ways to minimize the adverse effects.
2. Ways to include, in analysis and design of a space-robotic task, the effects of dynamic coupling between a robot and its supporting structure, and ways to minimize these effects.
3. Accounting of the effect of the initial configuration of a robot on the performance of a given task in an optimal manner.
4. What improvements in trajectory design for base reaction minimization could be achieved by using alternative cost functions and optimization schemes?
5. Could an algorithmic control approach such as adaptive control or nonlinear feedback control effectively solve the problems of base reaction minimization and disturbance (acceleration, jerk) limitation on payload?
6. How can more intelligent control approaches, knowledge-based control in particular, be employed to meet the performance objectives of a space robot?
7. How can the performance of a space robot be improved through the use of traction-drive joints? In fact these issues may not be limited to space robotics and can have implications in other applications such as industrial robotics.