17 March 2021 in IEEE Sensors Journal
In this paper we present a bionic multi-sensor anavigation and control system for unmanned ground vehicles to complete the homing task. The system consists of the pixelated polarized sensor, a Micro Inertial Measurement Unit (MIMU), and a monocular camera. To compensate for the installation error, we provide a joint calibration method for multiple sensors. Utilizing the measurements of pixelated polarized vision sensor, we firstly propose an adaptive integrated method with the Visual-Inertial System. The integrated algorithm can not only solve the ambiguity problem of polarized orientation and reduce the cumulative error of the system, but also increase the navigation output rate and enhance the robustness of the system. We present a homevector-based strategy for the goal-directed navigation and control of the unmanned ground vehicles. When the external data link is interrupted, the unmanned vehicles can return to the starting point autonomously, which benefits for improving the survivability of the system. Finally, we design various experiments to verify the algorithm proposed in this paper. The experimental results of the calibration demonstrate that the RMSE of the orientation after calibration is only 0.014°. In the navigation and homing experiment, the RMSE of the position error is 0.64m, and the minimum homing error is only 0.49m (1.09% of the travelled distance). Finally, we discuss interesting insights gained with respect to future work in multi-sensor integration and robot control strategies.
24 August 2022 in IEEE Sensors Journal
Skylight polarized patterns provide fresh thinking for the human to use the information sources in nature for orientation, but this orientation method is susceptible to weather conditions. Especially, in the case of cloudy weather, the cloud directly leads to the decline of orientation precision. In this article, a robust bionic polarized skylight orientation method based on the model consistency of polarization patterns (MCOPPs) is proposed for cloudy weather. First, we made full use of the DOP information to preprocess the polarization information as an aid to the subsequent polarization light orientation. Then, based on the vertical feature of the E-vector of the polarized light and the scattering plane, we developed the MCOPP error criterion and proposed a two-step optimizing framework for the effective selection of the interior points. Finally, the outdoor static rotating experiment and the dynamic vehicle experiment are conducted for evaluating our algorithm. The experimental results demonstrate that MCOPP can realize an effective and robust polarized orientation under cloudy conditions, and the accuracy of orientation in the outdoor static experiment is within 0.5° and in the dynamic car experiment is within 0.7°. Compared with the other four classic methods, the RMSE of the orientation error can be decreased by 50.6% in the outdoor static experiment and by 46.0% in the dynamic vehicle experiment. Especially, the proposed MCOPP has lower computational complexity than the existing robust polarized skylight method.
17 January 2024 in IEEE Transactions on Instrumentation and Measurement
In instances where satellite signals are obstructed or unavailable, long-endurance unmanned aerial vehicles face substantial navigational challenges. In an attempt to address this challenge, a biologically inspired global polarization positioning technology has been reported. While this technology offers a possible solution for autonomous global positioning, its practical application faces challenges due to the current limitations in positioning accuracy. Therefore, this paper introduces an iterative joint global polarization position strategy based on image feature and polarization pattern, aiming to achieve enhanced precision in global positioning. Firstly, a comprehensive polarization positioning model is constructed and the technical challenges are revealed. Subsequently, a solar image position solution method based on joint polarization pattern and image gray value is proposed, which can accurately calculate the solar image position in a single frame. Finally, a solar position iterative solution method is proposed and verified through outdoor experiments. Experimental results of more than 9 hours in 6 cities show that the proposed method improves the current polarization positioning accuracy, and the root mean square error of global positioning is 25.43 km. Overall, this model and algorithm have demonstrated the ability to accurately calculate the latitude and longitude of the carrier in real-time, while exhibiting strong adaptability to diverse environmental conditions.
14 April 2025 in IEEE Transactions on Aerospace and Electronic Systems
Skylight polarization is a significant physical characteristic of Earth's atmosphere. Inspired by the biological use of polarized skylight for navigation, polarization heading provides a reliable solution for autonomous navigation in environments where satellite navigation is denied. However, due to high computational complexity, polarization heading still faces challenges in dynamic experiments. This paper presents a fast and polarized heading detection algorithm that utilizes solely the additive property of polarization angles, ensuring its effectiveness and speed under clear, open sky conditions. Firstly, the additive criterion of the zenith is presented and theoretically proven. The mean value of the polarization angles at measurement points within a circular area centered at the zenith equals the solar azimuth angle. This criterion facilitates fast polarization heading. Subsequently, a method for analyzing the complexity of polarized heading is proposed, applicable to current polarization heading algorithms. The analysis indicates that the complexity of the existing polarized heading algorithms has been reduced from O(n2) and O(n3) to O(n). Finally, the heading of the carrier is determined by selecting the region that satisfies the additivity criterion based on the attitude angles. The experiment was conducted under open-sky conditions using a self-developed prototype of a polarized skylight compass, including simulation, static rotation experiments, and dynamic experiments. The experimental results demonstrate that, compared to classical algorithms, the proposed algorithm in this paper can reduce the time cost by up to 95%. Particularly in dynamic experiments, the algorithm processes each frame in less than 0.02 second without compromising heading accuracy, further enhancing its potential for application on a quadrirotor drone.
6 June 2025 in IEEE Robotics and Automation Letters
Visual-inertial odometry with low-cost visual and inertial sensors can provide accurate navigation results in GNSS-denied environment. However, the mainstream methods currently suffer from the inevitable accumulation of position and heading errors over long-term operation. Inspired by the compound eyes of insects such as ants, polarization skylight compass can obtain the accurate orientation from the sky polarization pattern without cumulative error. This letter proposes a complete polarization-based visual-inertial odometry (POL-VIO) for challenge environment. This system offers a visual-inertial odometry enhanced by a skylight compass, which reduces heading drift and provides high-accuracy state estimation under complex environment. Additionally, it achieves online spatio-temporal calibration of the polarization light compass, incorporating ambiguity correction and horizontal angle compensation in polarized light orientation. Besides, we utilize the degree of skylight polarization to determine whether compass measurements are reliable. The polarization skylight compass enhances the reliability of heading constraints in complex environments by utilizing the Angle of Polarization information. Results from real-world experiments substantiate that our system can achieve a notable improvement in positioning and orientation accuracy under significant obstructed sky observations after long time of operation. On three datasets with a total length exceeding 7 km, the average positioning error of POL-VIO is less than 2 m/km.