Fastening tool tracking system using a Kalman filter and particle filter combination

Abstract

This paper presents a position tracking system which estimates the position of the tip of a fastening tool. The proposed system uses a Kalman filter (KF) and particle filter (PF) combination to synthesize measurements from an inertial measurement unit (IMU) and a position sensor. The KF part is used to estimate the position of the centre of mass of the tool, and the PF is used to estimate the orientation of the tool. In addition, a rule-based logic system is used to reduce angular velocity measurement error and identify the fastening action of the tool. The proposed system was validated experimentally using various scenarios representative of assembly tasks in a factory environment. The experiment results show that the proposed system can accurately identify the fastened bolt even when the angular velocity measurement is not accurate provided that a large enough number of particles is used. In addition, even when there are multiple possibilities for fastened bolt positions, the experimental results show that the proposed system can correctly identify the fastened bolt by utilizing the accumulated position error of each particle.