Research on algorithms for estimating kinematic parameters of a system of articulated rigid bodies based on vectors of accelerations and angular velocities of selected ones

Determination of kinematic parameters of articulated rigid bodies is particularly important when recreating the trajectory of movement of given systems: human (running, skiing), machines (reproducing robot motion or dedicated mechanisms). Kinematic parameters such as rotation angle between members, angular velocity and linear acceleration are determined based on measurements from IMU (Inertial Measurement Unit) sensors that are mounted on individual members. In applications, the goal is to minimize IMU elements, which e.g. can hinder human movement. In this paper, Kalman Filter estimation was used to minimize the number of IMU sensors. The model of the system was prepared on the basis of Newton-Euler equations and Denavit-Hartenberg notation. The experiments included a 3-articulated pendulum model and the following three cases: i) placing IMU sensors on all three system segments, ii) on two segments, iii) on one system segment. Removed IMU sensors were replaced by estimates based on measurements of the segment preceding the examined element. The measurement error was verified using the reference measurement with the VICON vision system.