Toward recognition of object shape on a distributed wheeled array by using a sparse infrared sensor array

This paper presents an object shape estimation algorithm using a sparse IR sensor array which is installed in a wheeled array system. The system is in a modular shape and each module consists of three omni wheels and an IR sensor in a triangular configuration. A flat object with unknown shape is placed on top of the wheeled array and it is transported in a certain way that the sparse sensor array detects the shape of the object progressively. The algorithm uses an occupancy grid structure to represent and update the object shape information. IR sensor reading is modeled by Gaussian and the object shape is probabilistically updated based on the sensor reading and motion prediction. Initial results from experiments are presented, and future research topics are discussed.