A Robot Kinematics Model Estimation Using Inertial Sensors for On-Site Building Robotics

Abstract

In order to make robots more useful in a variety of environments, they need to be highly portable so that they can be transported to wherever they are needed, and highly storable so that they can be stored when not in use. We propose "on-site robotics", which uses parts procured at the location where the robot will be active, and propose a new solution to the problem of portability and storability. In this paper, as a proof of concept for on-site robotics, we describe a method for estimating the kinematic model of a robot by using inertial measurement units (IMU) sensor module on rigid links, estimating the relative orientation between modules from angular velocity, and estimating the relative position from the measurement of centrifugal force. At the end of this paper, as an evaluation for this method, we present an experiment in which a robot made up of wooden sticks reaches a target position. In this experiment, even if the combination of the links is changed, the robot is able to reach the target position again immediately after estimation, showing that it can operate even after being reassembled. Our implementation is available on https://github.com/hiroya1224/urdf_estimation_with_imus .

Figures (9)

• Figure 1: Sequence of kinematic model estimation. In an experiment to evaluate our method, we used centrifugal force to estimate the length of a link by shaking a manipulator made of wooden sticks. By using the link length estimation results, the inverse kinematics of the gripper can be solved toward the target position.
• Figure 2: Flowchart of our proposed relative pose estimation method. Corresponding sections are shown in each box.
• Figure 3: Time interpolation using Savitzky-Golay filter with $N=2$. If we want to align with the value at time $T_\text{interp}$, we can use polynomial interpolation to estimate the value at that time. If $T_\text{interp}$ falls outside the interpolation range, the closer of the values of $T[k \pm N]$ is used instead.
• Figure 4: Setup of the validation experiment. Two M5Stack Fire were used as the IMU modules. The IMU modules were fixed to aluminum rods with 200 mm spacing on the rods.
• Figure 5: Results of validation experiments. From the top row, the time variation of the angular velocity during the experiment, the estimated relative position, the position error, the orientation error, and the square root of the trace of covariance of the estimation results are depicted.