ABSTRACTIn this paper, a novel Automated Vehicle-Based Objective Tyre Testing (AVBOTT) concept is introduced. In brief, AVBOTT pursues the development of inexpensive automated testing vehicles capable of executing a set of standardised manoeuvres and generate large amounts of tyre data rapidly and inexpensively. Naturally, the costs associated with current vehicle instrumentation and, in particular, with Wheel Force Transducers (WFT) prevent this concept from realisation at the moment. In order to overcome this limitation, a cost-effective alternative to WFT is proposed in this paper to estimate the tyre forces using indirect measurement techniques. Specifically, for the first time, the validity of random-walk virtual tyre force sensors for vehicle-based objective tyre testing is studied comprehensively. Moreover, the suitability of Adaptive Neuro-Fuzzy Inference Systems (ANFIS) to extract relevant tyre attributes in a systematic manner from lateral and longitudinal dynamics manoeuvres is analysed in detail. The proposed solutions are assessed in a complete tarmac tyre testing program carried out at IDIADA with a fully instrumented experimental vehicle. Additional winter testing activities performed at Revi facilities (Sweden) complete the experimental evaluation at different mu levels.