Online Slip Detection and Friction Coefficient Estimation for Autonomous Racing

Abstract

Accurate knowledge of the tire-road friction coefficient (TRFC) is essential for vehicle safety, stability, and performance, especially in autonomous racing, where vehicles often operate at the friction limit. However, TRFC cannot be directly measured with standard sensors, and existing estimation methods either depend on vehicle or tire models with uncertain parameters or require large training datasets. In this paper, we present a lightweight approach for online slip detection and TRFC estimation. Our approach relies solely on IMU and LiDAR measurements and the control actions, without special dynamical or tire models, parameter identification, or training data. Slip events are detected in real time by comparing commanded and measured motions, and the TRFC is then estimated directly from observed accelerations under no-slip conditions. Experiments with a 1:10-scale autonomous racing car across different friction levels demonstrate that the proposed approach achieves accurate and consistent slip detections and friction coefficients, with results closely matching ground-truth measurements. These findings highlight the potential of our simple, deployable, and computationally efficient approach for real-time slip monitoring and friction coefficient estimation in autonomous driving.

Figures (5)

• Figure 1: Top view of the vehicle plant $P$ with a body-fixed 3D coordinate system $\{X,Y,Z\}$ attached at the center of mass. The figure illustrates the $X$-$Y$ plane, where $l_f$ and $l_r$ denote the distances from the CoM to the front and rear axles, and $l_w = l_f + l_r$ is the wheelbase. Forces $F_x$ and $F_y$ act at the CoM, $\delta$ is the steering angle, $\alpha$ is the slip angle, and $\psi$ is the yaw angle.
• Figure 2: Local body-fixed 3D coordinate system $B_w=\{X_w,Y_w,Z_w\}$ attached at the tire center. The slip angle $\alpha$ is defined as the angle between the tire longitudinal axis $X_w$ and the tire center velocity vector $V_w$.
• Figure 3: Map of the first, second, and third closed-loop tracks used for testing.
• Figure 4: Friction circle visualization for cardboard, acrylic, and tile surfaces. The blue dots denote measured longitudinal and lateral accelerations (in g); the star indicates the maximum traction point corresponding to the estimated friction coefficient $\mu$.
• Figure 5: Comparison of ground truth (gray) and estimated (blue) friction coefficients (µ) on tile, cardboard, and acrylic.

Theorems & Definitions (11)

• Definition 1: Plant
• Definition 2: State
• Definition 3: Sensors
• Definition 4: Controller
• Definition 5: Kinematic model
• Definition 6: Force
• Definition 7: Dynamical model
• Definition 8: Pure rolling
• Definition 9: Slip
• Definition 10: Traction coefficient