Research status and prospects of road surface type recognition and tire-road friction coefficient estimation

Road surface condition estimation is a critical technology for enhancing vehicle safety and autonomous driving performance. It comprises two main components: road surface type recognition and tire-road friction coefficient (TRFC) estimation. Road surface classification provides prior knowledge for friction estimation, while TRFC estimation serves to validate classification results. For autonomous vehicles, accurate road type recognition and friction coefficient estimation form the fundamental basis for safe navigation and decision-making. However, existing methods for road condition estimation still face challenges in terms of accuracy and real-time performance, particularly under extreme scenarios involving abrupt changes in road conditions. This paper reviews current domestic and international methods for road surface condition estimation based on their underlying principles. It introduces a novel classification perspective on TRFC that differs from traditional approaches. Furthermore, it analyzes emerging methodologies enabled by neural networks and vehicle-to-everything (V2X) technologies, and presents an outlook on future research directions in this domain.