Wheel Odometry Aided Visual-Inertial Odometry in Winter Urban Environments
Bibliographic record
Abstract
Over the last decade or so, the world has witnessed the rapid changes in the way people drive. How to ensure the navigation performance in challenging environments such as complex urban canyon environments or winter road environment with a relatively low-cost navigation system has become a popular research topic. Global Navigation Satellite System (GNSS) positioning is commonly used for land vehicle navigation. However, the accuracy of GNSS positioning is reduced in such challenging environments due to obstructions and multipath effects. Thus, the development of an alternative, accurate, inexpensive, and self-contained land vehicle navigation systems to bridge the GNSS gaps is significant for land vehicle navigation systems. Visual-inertial odometry (VIO) is an accurate, inexpensive, and complementary approach for land vehicle navigation in GNSS signal-denied environments. VIO is subject to scale drift because it estimates forward direction translation using distant feature points that are generally located only in the forward direction. Wheel odometer measurements can be obtained from the CANBUS interface of most modern passenger vehicles and these provide reliable estimates of the forward wheel speed. In this thesis, an innovative approach to incorporate wheel odometry (WO) and non-holonomic constraints (NHC) together with tightly-coupled monocular visual-inertial odometry using the Multi-State Constraint Kalman Filter (MSCKF) is proposed and implemented. The algorithm is first validated using the public KITTI Dataset [1] with simulated wheel odometer data. Then, the KAIST Complex Urban Dataset [2] is used to test the performance of IMU+Vision+WO integration system in urban canyon environments. Winter driving data is collected in Calgary and used to evaluate the influence of winter road conditions on the proposed algorithm. The results demonstrate that WO and NHC are able to control the scale drift, and as a result are able to control both scale and orientation over longer periods than IMU+Vision alone. IMU+Vision+WO achieved 1.814 m horizontal position error in a 1-minute drive in an urban canyon environment in the KAIST Complex Urban Dataset and 19.649 m and 3.456 m horizontal position errors in two 1-minute drives in our Calgary winter urban environment. The results demonstrate that IMU+Vision+WO is a very promising method to bridge the GNSS outages and performs very well in some challenging environments.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".