Current Non-Contact Road Surface Condition Detection Schemes and Technical Challenges

Abstract

Given the continuous improvement in the capabilities of road vehicles to detect obstacles, the road friction coefficient is closely related to vehicular braking control, thus the detection of road surface conditions (RSC), and the level is crucial for driving safety. Non-contact technology for RSC sensing is becoming the main technological and research hotspot for RSC detection because of its fast, non-destructive, efficient, and portable characteristics and attributes. This study started with mapping the relationship between friction coefficients and RSC based on the requirement for autonomous driving. We then compared and analysed the main methods and research application status of non-contact detection schemes. In particular, the use of infrared spectroscopy is expected to be the most approachable technology path to practicality in the field of autonomous driving RSC detection owing to its high accuracy and environmental adaptability properties. We systematically analysed the technical challenges in the practical application of infrared spectroscopy road surface detection, studied the causes, and discussed feasible solutions. Finally, the application prospects and development trends of RSC detection in the fields of automatic driving and exploration robotics are presented and discussed.

Lightweight Regression Model with Prediction Interval Estimation for Computer Vision-based Winter Road Surface Condition Monitoring

“…Classification-based studies typically cluster the road surface conditions to a single digit number of clusters, e.g. in the review of Ma et al [3] a total number of seven clusters were identified. Since the MAE and RMSE values scored by the networks were roughly around 10% of the total range of the friction factor value space, the errors were on average smaller than the resolution of classification approaches in general.…”

“…They offer alternative ways for the measurement, or they could be fused with contact-based methods. Review of non-contact friction estimation has been presented by Ma et al [3]. The non-contact approaches can be roughly divided into methods that utilise special optical sensors, and methods which utilise computer vision algorithms to process images captured with traditional visible light cameras.…”

“…Related code is available at https://github.com/ojalar/SIWNet Road surface condition can be analysed with several onboard methods [2], [3]. A common approach has been to utilise wheel dynamics information for the estimation task.…”

Estimation of Road Wetness from a Passenger Car

“…In this the second definition will be used. Among recent reviews dedicated to technologies that can be applied to road runways [10][11][12], it has been found that infrared spectroscopy methods are the mo cient compared to other contactless technologies. Despite their many advantages accuracy, reliable recognition of various contaminants, etc.…”

“…Recently, Riahi and coauthors [9] dev a similar system and, using a trailer equipped with a passenger car's tire, found pro relationships between the acceleration amplitude (root mean square of an accele time signal) and the water depth. Among recent reviews dedicated to technologies that can be applied to roads and runways [10][11][12], it has been found that infrared spectroscopy methods are the most efficient compared to other contactless technologies. Despite their many advantages (high accuracy, reliable recognition of various contaminants, etc.…”