Research and develop low-cost, soft robotic grippers having built-in actuation, sensing and embodied intelligence that enable safe-grasping, adaptability to object shape, and grasping versatility for reliable and efficient picking of mushrooms, and secondarily, of other agricultural produce with similar handling dynamics such as soft fruit (kiwifruit, berries, grapes etc.).
Research and develop the engineering and blending of novel materials that offer precise tuning of important material properties, safe interaction with the food elements (i.e. food-safe materials), have minimum impact on the environment (recyclable) and provide robust and maintenance-free production over a many cycles of operation (i.e. possess self-repair properties).
Research and develop a set of accelerated continuum mechanics modelling algorithms that will facilitate sophisticated model-based control schemes, capable of being executed by limited computational resources. This will be of paramount importance in achieving the complex control strategies that are required to reliably pick and outroot fresh mushrooms.
Research and develop advanced cognition capabilities of the soft gripper through a learning by demonstration framework comprising cutting-edge multi-task and meta learning techniques. Such techniques will enable a novel approach where force and torque profiles captured by a smart glove during human demonstration of mushroom picking will be extracted into a skill-capturing policy centered around individual goals.
