Investigating the Effects of Cooperative Driving for CAVs in Different Driving Scenarios Using Multi-Driver Simulator Experiments
This project aims to develop cooperative driving strategy (CDS) in a CAV environment and evaluate its safety effects. UCF’s multi-driver simulator will be used to collect driver interaction data while CARLA (an open-source autonomous driving simulation software) and other micro-simulation tools will be used to develop driver interaction scenarios. Both the freeway exit scenario and non-signalized intersection will be tested in the experiment. For the freeway exit scenario, Drivers will drive in a platoon and be interfered by a vehicle that traverses across lanes to leave the freeway; in addition, drivers will also play the role of leaving vehicle. For non-signalized intersection, drivers will be instructed to arrive at the intersection at a similar time from different approaches. Driver interaction data and safety indicators will be collected from three layers of experiment design. In the first layer, all involved drivers do not receive any CDS signal and this layer is the baseline. In the second layer, drivers are receiving CDS recommendations and need to determine driving strategies on their own. In the third layer, involved vehicles will be CAVs which are controlled by CDS; no driver operation will be needed in this layer.