Use of robotics within the industrial environment has allowed improved productivity, but only for aspects that are repetitive and thus programmable. Any deviation from a predefined situation causes the system to fail thus there is still the involvement of the human element. Bringing the strength of a robotic system, with the intelligence of a human controller allows increased adaptability. Industries where there is a variability of action but require additional strength benefit from this combination of man and machine.
Regarding this, there is a gap in regards to upper-body exoskeletons that enhance the user’s strength at full reach. Current systems augment the user’s carrying capacity, but do not state what their load manipulation at full reach.
The literature survey also shows that there is a gap in regards to optimisation of the exoskeleton systems with respect to both the geometry of the joints and the selection of the hydraulic circuit design. Current systems appear to have tried to get a prototype working as quickly as possible without simulation of the system first.
The first focus of this research is in the development of optimisation of joint geometry for revolute and gimbal joints. Several different designs are broken down into geometric equations that can then be fed into an optimisation routine to determine the ideal geometry for a given loading and motion range.
The second focus of this research is using the optimised joint geometry and design an upper-body exoskeleton that has load manipulation at full reach. This design consisted of an initial load manipulator, the elbow and the shoulder joint. This design was found to be heavy, but for similar load carrying methods of other exoskeletons and robotics, has a similar load to weight ratio.
The third focus of this research was a review of current hydraulic circuits in regards to an enhancive exoskeleton system. previous exoskeletons did not focus on this aspect of the design, meaning that efficiency and optimisation options have not been explored. Though servo valve systems can be used, additional benefits can be found with full pressure override regeneration circuits, so that return fluid can be used productively. Though pump based circuits were found not to be suitable for this design, they have shown promising energy recovery options for lower power systems.
The final focus of this research was to bring together the upper-body structural design and selected hydraulics and simulate them in a virtual environment. This saves money on component manufacture and has shown that there are issues to overcome in future development. The frequency response of the valves and system were found to be lower than required to follow the human motion which would be a limiting factor for zero loading on the user. The motion capture was also not optimised to match the motion limitations of the exoskeleton resulting in the actuators reaching the end stop limits.
The final conclusions of this research are that the optimisation design routines allows the easy selection of actuators and geometry to avoid singularity events in the motion and that hydraulic regeneration circuits will give a benefit to exoskeleton motion. The use of simulation over a prototype has meant that the costs have been reduced, and outlined areas of concern without wasted manufacturing.
