Composite Learning Adaptive Safety Critical Control With Application to Adaptive Cruise of Intelligent Vehicles

This article presents an adaptive safety critical control scheme for uncertain systems with potentially conflicting control objective and safety constraint. The modified control barrier function (MCBF) is presented to rigorously guarantee the safety constraint subject to the unavoidable parameter estimation errors, then quadratic program (QP) is employed to synthesize the control Lyapunov function (CLF) and MCBF to form the certainty equivalence controller. Since the priority of CLF is regraded and the MCBF is designed to be nonpositive definite, we employ a separate parameter update module rather than the Lyapunov-based adaptive control approaches. The composite learning method is presented to eliminate the effects of parametric uncertainties without imposing the restrictive excitation conditions, further to avoid the conservative control performance. The experimental results of adaptive cruise for intelligent vehicles are provided to verify the theoretical findings.