Safety-Guaranteed Driving Control of Automated Vehicles via Integrated CLFs and CDBFs

To achieve certain automated driving control with guaranteed driving safety, such as an automatic lane change with a high speed on a low friction road (for emergent obstacle avoidance), an appropriate vehicle control design satisfying multiple safety constraints is required. An integration of control Lyapunov functions (CLFs) with control barrier functions (CBFs) is an effective method to simultaneously achieve good tracking control performance and satisfy safety constraints. However, for automated vehicle (AV) control, CBF may not be applicable since some AV safety constraints are time-varying and/or control-dependent. To handle the control-dependent and time-varying safety constraints, control-dependent barrier functions (CDBF) and time-varying CBFs (TCBFs) are integrated with CLFs in a quadratic programming problem, which aims to achieve good tracking control with guaranteed vehicle safety. In the integration, the mismatch between different forms of constraints is coordinated by using the second-order invariant constraints. To verify the effectiveness of the proposed methods, CarSim®/Simulink co-simulations were conducted and comparisons are presented and discussed.