Targeted potential project (2023)

Huawei collaboration with FDE ITF? still under disucssion

Abstract

Accurate and safety-quantifiable localization is of great significance for safety-critical autonomous systems, such as unmanned ground vehicles (UGV) and unmanned aerial vehicles (UAV). The visual odometry-based method can provide accurate positioning in a short period but is subject to drift over time. Moreover, the quantification of the safety of the localization solution (the error is bounded by a certain value) is still a challenge. To fill the gaps, this paper proposes a safety-quantifiable line feature-based visual localization method with a prior map. The visual-inertial odometry provides a high-frequency local pose estimation which serves as the initial guess for the visual localization. By obtaining a visual line feature pair association, a foot point-based constraint is proposed to construct the cost function between the 2D lines extracted from real-time images and the 3D lines extracted from the high-precision prior 3D point cloud map. Moreover, a GNSS receiver autonomous integrity monitoring (RAIM) inspired method is employed to quantify the safety of the derived localization solution. Among that, an outlier rejection (also well-known as fault detection and exclusion) strategy is employed via the weighted sum of squares residual with a Chi-squared probability distribution. A protection level (PL) considering multiple outliers is derived and is utilized to quantify the potential error bound of the localization solution in both position and rotation aspects. The effectiveness of the proposed safety-quantifiable localization system is verified using the datasets collected in the UAV indoor and UGV outdoor environments.

Dr. Wen, W. and Zheng Xi.

Scholarship Plan

• (1st year) 09/2022-08/2023: ZGD2
• (2nd year) 09/2023-08/2024: University Central Fund (3+1)
• (3rd year) 09/2024-08/2025: University Central Fund (3+1)

Recent Research Plan: Xi Zheng (Huawei ZGD2+ “PolyU 3+1”)

• Half year research plan (by Dec 2022)-IEEE T-ITS
  • Visual localization in prior map with line model
  • PL estimation with multiple faults
• 0.7 year research plan (by March 2023)-IEEE ITSC conference
  • Study the tightly-coupled Integration of the Visual/inertial and Line-map matching
  • Study how the PL is decreased compared with the soly line based method
  • Dr. Hsu suggest to focus more on PL theory
• One year research plan (by Jun 2023)-IEEE T-ITS
  • Visual localization in prior map convexity quantification
  • Visual outlier rejection with multiple-level optimization
• 1.5 years research plan (by Dec 2023)- IEEE T-ITS
  • Safety constrained visual localization
  • Visual localization error modeling with GMM and its safety quantification

Student Supervision

• Supervise Edward? (undergraduate student)

News

• [02/07/2023] First discussion in new year 2023

• [12/01/2022] We are going to design the simulated dataset from Carla to validate the proposed safety quantification method.

• The manuscript entitled “ Safety-quantifiable Line Feature-based Monocular Visual Localization with 3D Prior Map “ is submitted to the IEEE Transactions on Intelligent Transportation Systems. The video about this paper is avaiable by Youtube and Bilibili.

• Discussion on 02th, Nov 2022
  • Revise the manuscript based on Dr. Hsu’s comments and suggestions
  • Read the paper suggested by Dr. Hsu
    • Reid, T. G., Houts, S. E., Cammarata, R., Mills, G., Agarwal, S., Vora, A., & Pandey, G. (2019). Localization requirements for autonomous vehicles. arXiv preprint arXiv:1906.01061.
    • Weighted RAIM for Precision Approach
    • https://saemobilus.sae.org/content/12-02-03-0012/ (Localization Requirements for Autonomous Vehicles)
    • Add the discussion/comparison between the proposed and the slope based PL evaluation
  • Expected to get the revised version from Zhengxi by next week (9th Nov 2022)
• Nov 2022, 1 paper is going to be submitted to IEEE Transactions on Intelligent Transportation Systems.
  • Zheng, X., Wen, W., Hsu, L.-T.: Safety-quantifiable Line Feature-based Monocular Visual Localization with 3D Prior Map for Autonomous Systems. IEEE Transactions on Intelligent Transportation Systems. 2022