Crowdsourcing Radar Maps with AUTO’s Integration of Multiple Imaging Radars and INS/GNSS for Autonomous Applications
Bibliographic record
Abstract
Enabling reliable and accurate localization in all environments typically requires the fusion of information from many different sensors and sources. One common approach is the fusion of perception sensors and high definition (HD) maps. However, obtaining HD maps is often a challenge, due to the high financial costs, large data requirements, and heavy processing demands that are involved. This paper presents AUTO, a real-time integrated navigation system that implements a novel method for map crowdsourcing using imaging radars. AUTO integrates inertial navigation systems (INS), global navigation satellite systems (GNSS), odometer, and multiple radars sensors to crowdsource radar-based maps of the environment. Data is gathered by the same systems used for real-time positioning, thereby eliminating the need for costly survey equipment. Furthermore, large maps are divided into smaller areas using a tiling scheme to limit memory growth and improve processing requirements for map-building. This approach allows the map boundaries to be automatically determined based on the input data. The results demonstrate the accurate mapping of downtown areas in Detroit, MI, and Calgary, AB, using combinations of 3 and 5 imaging radars. The maps are represented as 2D occupancy grid maps generated from real-world data, with a 10cm resolution. The results also show how AUTO can then use the crowdsourced radar maps for reliable and accurate positioning in challenging environments. Key performance indices (KPI) are presented for vehicle using different multi-radar configurations. The presented solution also features integrity monitoring for the integrated navigation solution with protection levels illustrated using a Stanford diagram. AUTO was tested under a wide variety of environments, locations, lighting, and weather conditions to assure the robustness and reliability required by autonomous applications.
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 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.000 | 0.000 |
| 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".