Experimental investigation of a maneuver selection algorithm for vehicles in low adhesion conditions

TL;DR

This work tackles safe vehicle maneuvering under low-adhesion conditions by coupling a model-based estimator with a data-driven maneuver selector. The estimator outputs a compact set of physics-ground parameters, including the inertial velocity $v$, friction coefficient $μ$, cohesion $c$, and internal shear angle $φ$, from high-dimensional sensor data, which feed a predictor trained to choose among five discrete maneuvers. Experimental validation on a 1:5 scale platform demonstrates real-time parameter estimation, rapid convergence to ground-truth values, and successful maneuver optimization that adapts to hard and deformable terrains. The approach offers a practical pathway to improve winter-road safety by enabling reactive, ground-aware maneuver selection with real-time capability and limited data requirements.

Abstract

Winter conditions, characterized by the presence of ice and snow on the ground, are more likely to lead to road accidents. This paper presents an experimental proof of concept, with a 1/5th scale car platform, of a maneuver selection scheme for low adhesion conditions. In the proposed approach, a model-based estimator first processes the high-dimensional sensors data of the IMU, LIDAR and encoders to estimate physically relevant vehicle and ground conditions parameters such as the inertial velocity of the vehicle $v$, the friction coefficient $μ$, the cohesion $c$ and the internal shear angle $φ$. Then, a data-driven predictor is trained to predict the optimal maneuver to perform in the situation characterized by the estimated parameters. Experimental results show that it is possible to 1) produce a real-time estimate of the relevant ground parameters, and 2) determine an optimal maneuver based on the estimated parameters between a limited set of maneuvers.

Figures (12)

• Figure 1: Proposed approach for the maneuver selection algorithm.
• Figure 2: Photo of the modified Traxxas X-MAXX used for the experimentation.
• Figure 3: Illustration of the slip-slope relation for typical flat grounds.
• Figure 4: Illustration of the model-based estimator.
• Figure 5: Illustration of the discrete set of maneuvers, where (1) is brake 100 %, (2) is brake ABS, (3) is turn 100 %, (4) is turn 100 %, brake 100 % and (5) is turn 100 %, brake ABS.