Source-linked AI summary
Beyond the Proving Ground: Independent Public-Road Testing of Assisted Lane Change Systems using LiDAR
Marcello Cellina, Akos Kriston, Antonio Migneco, Davide Maggi, Stefano Favelli, Fabrizio Re, Fabrizio Minarini, Andrea Nuovo, Riccardo Dona, Biagio Ciuffo
TL;DR
Proving-ground testing may miss public-road complexity, and geo-fencing can limit independent assessment of commercial ALC systems. The paper presents LiDAR-based public-road testing against R79 requirements and finds six completed manoeuvres below the prescribed distance, with three remaining significant after uncertainty is considered. The methodology is shown to support independent assessment under real-world conditions, though results are limited by testing a single vehicle and measurement precision.
Problem
Proving-ground scenarios may not represent real-world complexity, while geo-fencing can hinder independent testing and market surveillance.
Method
The paper uses LiDAR-based measurements to independently test commercial R79-approved ALC systems on public roads.
Results
6 of 27 experiments completed with an R79 Critical Distance Overshot, and 3 of 6 retained a 99% Confidence Interval measurement after uncertainty was considered.
Takeaways & Limitations
The methodology is precise enough to detect critical ALC behaviour and could support independent market surveillance and in-service monitoring under real-world conditions.
Takeaways & Limitations
The study is limited by testing a single vehicle and by measurement and temporal annotation precision.
Abstract
from arXiv · showhide
Testing of commercial Advanced Driver Assistance Systems is essential to ensure safety and compliance during type approval and in service operation. However, proving ground scenarios may not reflect real world driving complexity, while geo fencing can require manufacturer collaboration and limit assessment independence. This work presents a methodology for independently testing Assisted Lane Change systems on public roads. A campaign on the A31 French motorway used a test vehicle equipped with a LiDAR based vehicle detection and tracking system. Tests covered combinations of inter vehicle distance and speed between the test vehicle and the take over vehicle. Real time kinematic global navigation satellite system receivers assessed detection and tracking performance. Recorded lane change trajectories were compared with the lane change suppression requirements of UNECE Regulation Number 79. Of 27 predefined lane change manoeuvres, 18 were completed and 9 suppressed. In 6 cases, the system allowed manoeuvres that did not meet regulatory minimum distance requirements. In 3 cases, the deviation remained statistically significant after accounting for measurement uncertainty. To the authors knowledge, this is the first public road campaign designed to assess Assisted Lane Change compliance with Regulation Number 79 safety distance requirements. The results demonstrate the suitability of LiDAR based sensing for this purpose. The methodology can support market surveillance and future regulatory revisions by revealing real world behaviours not covered by approval procedures.
INTRODUCTION
Existing approval testing may not represent public-road complexity, while geo-fencing and GNSS constraints hinder independent assessment. This work addresses the gap with a LiDAR-based public-road methodology for independently evaluating commercial ALC systems against R79.
- Proving-ground testing is repeatable but covers limited scenarios that may not represent real-world driving complexity.
- Geo-fencing can require manufacturer collaboration, creating a barrier to independent market-surveillance testing.
- The study introduces a LiDAR-based methodology for independently testing commercial ALC systems on public roads instead of conventional proving-ground procedures.
- Prior studies address ALC development, field experiments, users, models, or broader safety assessment rather than dedicated R79 compliance testing of production systems.
- The literature lacks a public-road campaign explicitly evaluating ALC behaviour against R79 with vehicle-mounted LiDAR, including suppressed attempts.
METHODS
The method evaluates whether an approaching vehicle is sufficiently far away when the VUT crosses the lane marking. R79 defines this Critical Distance from vehicle speeds, deceleration, reaction time, and resulting time gap.
- The Critical Distance is the minimum separation between the VUT and approaching TO vehicle required before an ALC manoeuvre may start.
- The Critical Distance is defined as a function of the ALC vehicle and approaching vehicle speeds.
- a=3m/s^2, t_B=0.4s, and t_G=1s represent maximum approaching-vehicle deceleration, reaction time, and minimum resulting time gap, respectively.
- R79 evaluates the Critical Distance when the VUT crosses the lane marking at the Time of Start Measure.
- If the TO vehicle is closer than the Critical Distance at lane-marking crossing, the ALC system should suppress the manoeuvre.
Experimental Setup and Testing Procedure
The campaign independently tested a commercial ALC system on French motorways using coordinated VUT, TO, and support vehicles, LiDAR, cameras, and GNSS. A full matrix varied speeds and triggering distances, recording completed and suppressed manoeuvres for comparison with R79 critical-distance behaviour.
- The campaign drove three type-approved passenger cars on the northbound A31 motorway between Dijon and Nancy.
- The VUT carried a category C R79 ALC system, while the TO vehicle represented the approaching vehicle and carried rooftop LiDAR and cameras.
- The support vehicle maintained a fixed ACC headway behind the VUT and triggered manoeuvre starts at the desired distance.
- The LiDAR system provided tracked vehicle poses, while GNSS, cameras, and manual annotations supported position, lane-crossing, and indicator timing assessment.
- The test matrix combined VUT and TO speeds with three triggering distances selected around the R79 critical distance and minimum activation distance.
- 27 ALC tests produced 18 completed and 9 suppressed manoeuvres across 100–130 km/h speeds and 20–60 m triggering distances.
- Measurement and procedural limitations included non-constant activation delays, TO ACC reactions, limited LiDAR range, and three completed experiments outside LiDAR range.
- Six completed manoeuvres were classified as potential critical-distance overshoots, with 33% initiated below the R79 critical distance.
DISCUSSION
LiDAR measurements identified six Critical Distance Overshoots, with three remaining significant at the 99% confidence level after accounting for uncertainty. The overshoots occurred mainly at low speed differences, while none were judged dangerous by testing personnel.
- 0.83 m position precision and 1.40 km/h velocity-estimation precision characterized the LiDAR VDT output.The position and velocity precisions were measured against RTK GNSS intervals with RTK Fixed integer-level precision; lane-marking-crossing annotation had 1-second resolution.
- 3 experiments—16, 16b, and 16d—showed Critical Distance Overshooting at the 99% confidence level.Experiments 10, 10b, and 10d had only 2-sigma significance.
- 6 Critical Completed experiments violated the R79 Critical Distance boundary despite the ALC manoeuvre not being suppressed.The overshoots were identified from LiDAR measurements and categorized with measurement uncertainty considered.
- Low delta-v characterized all Critical Completed experiments in which the ALC was not suppressed despite an R79 Critical Distance Overshoot.The R79 boundary still imposes a 1-second headway when deceleration is not necessarily needed.
- 0 Critical Completed ALC manoeuvres were perceived as dangerous by the testing personnel.The identified overshoots occurred outside the operating-domain areas prescribed for R79 type-approval testing.
CONCLUSIONS
The paper presents an independent public-road methodology for assessing commercial R79 type-approved ALC systems using LiDAR. Applied to 27 experiments, it found six Critical Distance Overshoots, three significant at 99% confidence, and supports real-world market surveillance despite stated scope and measurement limitations.
- LiDAR-based public-road testing enabled independent assessment of commercial ALC systems without manufacturers’ support or high-precision RTK-GNSS dependence.The methodology targets geo-fenced systems type approved under UNECE Regulation No. 79.
- 27 ALC experiments used a full factorial design spanning different relative distances and speeds.Comparison with the R79 Critical Distance threshold identified six completed manoeuvres with Critical Distance Overshoots.
- 3 of 6 Critical Distance Overshoots remained significant at the 99% confidence level after spatial and temporal uncertainty was considered.
- Public-road LiDAR testing was precise enough to detect critical system behaviour in real-world operating regions insufficiently represented by conventional type-approval testing.The methodology could support independent market surveillance and in-service monitoring.
- Testing was limited by use of a single vehicle and by measurement and temporal-annotation precision.Future work is intended to improve LiDAR and annotation accuracy and develop a repeatable procedure.
DECLARATION OF CONFLICTING INTERESTS
The authors report no conflicts of interest and no financial support for the research, authorship, or publication.
- The authors declared no potential conflicts of interest regarding the research, authorship, or publication.
- The authors disclosed no financial support for the research, authorship, or publication.