Mobile robotics is an area that has seen explosive growth in the form of several research efforts in the recent past because of increasing computational power being available in small electronic packages and improvements in sensor technology. A mobile miniature device can benefit greatly by having its structure constructed with 'smart' materials. The device, during the course of its operation, is expected to encounter hostile terrain and environmental conditions. It is desirable that the materials and structures that compose the device respond to the environment appropriately and either avoid or minimize any damage. Smart structures have been developed that integrate fiber-optic sensors, piezoelectric actuators, and shape-memory-alloy fibers/rods that act as sensors and actuators. The scale associated with these materials/structures tends to be an order of magnitude larger than the envisioned mobile miniature device. A major challenge in this research effort is to scale down these materials, sensors, and actuators, and yet retain their functionality. Another challenge is to integrate multiple sensors and actuators within the same component/structure and coordinate their action to achieve multiple functionality.

A suitable miniature mobile robot capable of manipulation with a multi-link arm is an ideal solution to sensor deployment, adjustment, and replacement. Due to the small size of a miniature mobile robot, the manipulator must be light and relatively powerful to pick up and work with objects. A manipulator actuated by Shape Memory Alloy (SMA) wire is an excellent candidate to satisfy these requirements. However, SMA wires introduce further system nonlinearities and uncertainties, which lead to difficulties in accurate control of manipulation. Novel nonlinear control algorithms must be developed which are capable of accurately and efficiently positioning such a manipulator in any configuration.

The problem is the development of miniature mobile manipulators that are capable of sensing and actuation. The focus of the underlying research is to investigate kinematics, dynamics, control, design, sensors and actuation of a general purpose micro-robot that is able to go through a variety of smooth and rough terrains, avoid obstacles, and is also able to perform manipulation tasks (e.g., assemble, repair, install) once it reaches its destination. The manipulator may be a multi-link robotic arm actuated by SMA wire or servomotors. Comparison of a servo-actuated arm and SMA actuated arm will provide further insight for effective actuation.

The end goal of this project has been to build a miniature, multi degree-of-freedom robotic arm, powered by SMA actuators, and equipped with a miniature gripper for grasping and handling objects smaller than a ping-pong ball. The specific objectives of our team were to develop a set of miniature actuator systems that had the following features.

•   Small in size; approximately, 1000 cm³.
•   Mobile; the devices should be capable of motion outdoors in possibly damp and rough conditions. Should have the range of a football field.
•   Manipulation; the devices should be able to manipulate small objects (such as micro-sensors). Should have three to six degrees of freedom.
•   Capable of grasping, installing and adjusting sensor arrays.
•   Robust and reliable in operation.
•   Project Details
•   Research Issues
•   Project Publications