To address the lateral control issue between the automatic controller and the driver for lane changing assistance system, we present a novel multi-mode adaptive shared control strategy in optimal control framework. Different from traditional approaches that achieve shared control through adjusting control authority allocation, we aim at designing a human-friendly adaptive shared control scheme which can provide the driver with assistance in accordance with driving demands via multi-mode switching. Three control modes are proposed: 1. minimum-lateral error mode; 2. minimum-lateral acceleration mode; 3. minimum-interference mode. Meanwhile, we develop a hybrid model based on hidden markov model (HMM) and support vector machine (SVM) to recognize the optimum control mode. Mathematically, the shared control scheme is described as an optimal control problem, which is then addressed by using the improved Radau pseudospectral method (RPM). Numerical analysis and the driver-in-the-loop (DIL) tests show that the proposed shared control strategy has the advantage of ensuring driving safety while reducing driver’s workload.

为了解决车道变更辅助系统的自动控制器和驾驶员之间的横向控制问题，我们在最优控制框架中提出了一种新颖的多模式自适应共享控制策略。与通过调整控制权限分配来实现共享控制的传统方法不同，我们旨在设计一种人性化的自适应共享控制方案，该方案可以通过多模式切换根据驾驶需求为驾驶员提供帮助。提出了三种控制模式：1.最小横向误差模式；2.最小横向加速模式；3.最小干扰模式。同时，我们开发了基于隐马尔可夫模型（HMM）和支持向量机（SVM）的混合模型，以识别最佳控制模式。数学上 共享控制方案被描述为最佳控制问题，然后通过使用改进的Radau伪光谱方法（RPM）解决。数值分析和驾驶员在环（DIL）测试表明，所提出的共享控制策略具有确保驾驶安全，同时减少驾驶员工作量的优点。