Our goal is to design one 3D haptic sensation system covering all surfaces over the humanoid to help the robot to find its embodiment. Instead of applying dense array shaped sensors or monitoring the robot using multi external cameras, we propose to place as few sensors as possible to predict the touch information. In another word, we want to find the optimal positions all around the humanoid to place a few sensors optimally and sparsely, and then integrate with Machine Learning Model to predict the force information with high prediction accuracy.
Optimally and sparsely placed sensors are located internally considering system robustness. Rather than modifying the skin properties of the skeleton by attaching new skin, we propose to make use of the structure and material properties directly and measure the skeleton's deformation pattern using sparse sensor configuration.