Rehabilitation robotics is by nature an interdisciplinary field which combines aspects of industrial robotics, mobile robotics, teleoperation, software engineering, and human-machine interaction. In this research, design principles and guidelines for the structure of software systems were explored, and led to the development of a generic framework for the integration of shared control algorithms in a manipulator control system. Additionally, literature from the fields of teleoperation and mobile robotics was investigated in order to identify research areas in these fields which bear similarities to those of rehabilitation robotics.

Research in the field of human control of reaching and grasping movements was used as the basis for the design of sensor-based trajectory modification algorithms, and to guide the development of human-machine interaction strategies. The resulting shared control algorithms are described, and details of their implementation both in a simulated environment and in a hardware-based system are given. The hardware-based system was initially controlled through a DOS-based environment, then subsequently using the Windows-based Cambridge University Robot Language (CURL) environment. The incorporation of the trajectory modification algorithms with each of these manipulator-control programs is described.

The hardware-based system was evaluated in two sets of trials with able-bodied subjects. This provided a quantitative and qualitative assessment of the trajectory modification algorithms, and of three interactive, robot control strategies. The results of these evaluations are given, and showed that task performance could be improved through the use of the sensor-based trajectory modification algorithms. Finally, the overall conclusions and contributions of this thesis are discussed, and areas of this research which could benefit from further exploration are identified.

Theses