A Novel Control Framework of Brain-Controlled Vehicle Based on Fuzzy Logic and Model Predictive Control

Brain-controlled vehicles (BCVs) have vital practical values for the disabled and healthy people. To improve the performance of existing BCVs and lower the workload generated by BCVs to drivers, in this paper, we propose a novel control framework of BCVs, which consists of a brain-computer interface (BCI) with a probabilistic output model, an adaptive fuzzy logic-based interface model, and a model predictive control (MPC) shared controller. The BCI with a probabilistic output model can output all commands in a probabilistic form rather than a specific single command once. The adaptive fuzzy logic-based interface can convert the probabilities into the vehicle's input signals (including the vehicle acceleration and the increment of steering wheel angle) according to the vehicle state and road information. The MPC shared controller can ensure the control authority of brain-control drivers and reduce drivers' workload on the premise of maintaining safety. We establish an experimental platform to validate the proposed method by using the intersection selection and obstacle avoidance scenarios with eight subjects. The experimental results show the effectiveness of the proposed method in improving driving performance and decreasing drivers' workload. This work can contribute to the research and development of BCVs and provide some new insights into the study of intelligent vehicles and human-vehicle integration.