The principal objective of the research is to generate knowledge for use by systems engineers to develop machines for automated handling of a broad range of non-rigid materials.
Recently there have been a number of attempts to develop custom systems to automate specific tasks and it has been difficult to judge the extent to which these methods can be applied to other processes involving different materials.
This research programme has addressed the problem from a more general perspective. It has identified groups of handling tasks and non-rigid materials and has identified the generalised needs for automation technology leading to widely applicable new techniques to support the planning and implementation of automation systems for these materials. The research has produced new techniques in sensing, manipulation, gripping, simulation and knowledge base technologies for broad groups of common handling tasks dealing with materials from sheet to dough products.
The simulator is able to produce simulation of handling machinery integrated with the dynamic and static behaviour of the material. This can be used for planning automatic handling strategies and the configuration of complete handling systems. The knowledge base is able to automatically define suitable handling techniques for broad classes of material type and a range of handling tasks. This information is useful to control an automated system as well as for systems designers to prepare for future handling systems dealing with non-rigid materials. A system analysis has shown that the system can be realised as a real time system.
The studies have also lead to new sensing techniques able to identify material properties and to detect the behaviour of the material. Methods of automatically adjusting the manipulation or grip of the material to satisfy the requirements of the task have been demonstrated.
Categories will be defined through manipulation requirements, deformation tolerances and other qualities relating to handling machine system specification. Handling processes across the material range pastas, meats and textiles will be considered and a suitable process will be identified as a focus for detailed study. The studies will enable comparison of handling needs and likely solutions across the range of materials through construction of a demonstrator system integrating sensing, manipulator and pert system technologies.
The handling strategy employed by the system will depend upon task requirements and properties of the materials. Vision, force and torque sensing methods will be developed for identification of the properties of the non-rigid material strategies and an expert system will be developed relating the properties and requirements of the handling process to handling strategies.
Deformation reference models will be developed based on mechanical and physical properties, and schemes for the automatic selection and application of control strategies will be investigated using experimental and simulation methods. Sensing schemes to provide real-time feedback for purposes of automatic control of the manipulator and end-effectors will be investigated by experimentation.