Odometry Estimation for Aerial Manipulators
This chapter explains a fast and low-cost state localization estimation method for small-sized UAVs, that uses an IMU, a smart camera and an infrared time-of-flight range sensor that act as an odometer providing absolute attitude, velocity, orientation, a
- PDF / 763,919 Bytes
- 10 Pages / 439.37 x 666.142 pts Page_size
- 72 Downloads / 204 Views
Abstract This chapter explains a fast and low-cost state localization estimation method for small-sized UAVs, that uses an IMU, a smart camera and an infrared timeof-flight range sensor that act as an odometer providing absolute attitude, velocity, orientation, angular rate and acceleration at a rate higher than 100 Hz. This allows estimating almost continuously the localization of the aerial robot, when GPS or other methods can at most reach 5 Hz. This technique does not require creating a map for localization.
1 Introduction Combinations of Inertial Measurement Units (IMU) and monocular visual sensors for aerial robot localization are becoming very popular, thanks to their low weight, power consumption and cost, and their ease of installation. This constitutes a minimalist yet powerful sensor suite for autonomous localization as it allows recovering both the high motion dynamics and the localization with respect to the environment, including scale and, most important for aerial robot navigation, the direction of gravity. The processes of estimating the vehicle state using such sensors is either known as Visual-Inertial Odometry (VIO, with no absolute localization) [1], or as VisualInertial SLAM (enabling absolute localization by revisiting previously mapped areas) [2–4]. During an aerial manipulation mission, the focus of our work is not at building
A. Santamaria-Navarro (B) NASA-JPL, California Institute of Technology, Pasadena, CA 91109, USA e-mail: [email protected] J. Solà · J. Andrade-Cetto CSIC-UPC, Institut de Robòtica i Informàtica Industrial, Llorens i Artigas 4-6, 08028 Barcelona, Spain e-mail: [email protected] J. Andrade-Cetto e-mail: [email protected] © Springer Nature Switzerland AG 2019 A. Ollero and B. Siciliano (eds.), Aerial Robotic Manipulation, Springer Tracts in Advanced Robotics 129, https://doi.org/10.1007/978-3-030-12945-3_15
219
220
A. Santamaria-Navarro et al.
maps but at localizing the platform, thus we concentrate on VIO. The methods described in this chapter summarize the work presented in [5, 6].
2 Flow-Inertial-Range Odometry Here, we present a fast and low-cost state estimation method for small-sized UAV. We use exclusively low-cost sensors and low-complexity algorithms. As hardware, we take advantage of a low-cost IMU, a smart camera [7] which directly outputs 2D flow, and an infrared time-of-flight range sensor [8], all shown in Fig. 1. As software, we have developed two Kalman filters, in the extended (EKF) and error-state (ESKF) forms [9], together with a wide range of variations in the inner details, for the sake of performance evaluation and comparison. The overall estimation system acts as an odometer that provides absolute altitude, velocity, orientation, angular rate, and acceleration, with respect to a precise gravity reference, at a rate of 100 Hz. The x and y positions and the yaw angle are not observable, and their output is the result of an incremental estimation subject to drift—these modes can be observed with a lower update rate by a high
Data Loading...
