基于双层驾驶员模型的履带车辆纵向与横向协同跟踪控制方法

为解决基于离合器转向机速差转向履带车辆的路径跟踪控制问题，采用一种基于双层驾驶员模型的纵向与横向协同跟踪控制方法，以实现对期望路径的跟踪控制。第1层驾驶员模型以高斯混合隐马尔可夫模型为基础，以期望跟踪轨迹为输入，实现对转向模式序列的预测输出及切换控制，其中转向模式包括直驶模式、行进间转向模式与原地转向模式；第2层驾驶员模型以模糊逻辑来表征行进间转向模式下经验驾驶员纵向与横向控制配合规律，并最终依据期望航向校正偏差生成纵向、横向控制量。试验结果表明，上述控制算法能够有效地利用驾驶员的先验经验，解决车辆纵向与横向强耦合特性及转向不确定性给跟踪控制系统带来的挑战，最终实现车辆在特定场景下轨迹跟踪误差小于1.0 m的路径跟踪控制。

Longitudinal and Lateral Path Following Coordinated Control Method of Tracked Vehicle Based on Double-layer Driver Model

A longitudinal and lateral joint path following coordinated control method based on double-layer driver model is proposed to solve the path following control problem of tracked vehicle with clutch and brake steering structure. The first layer of driver model is based on the Gaussian mixture model and hidden Markov model. The desired path is taken as an input. The aim is to predict the steering mode sequences and switch the control mode according to the sequences. The second layer of driver model is established by using the fuzzy logic to reflect the skilled driver’s longitudinal and lateral operation experience. The longitudinal and lateral control commands are calculated according to the desired deviation of course. The results show that the driving experience can be utilized effectively to settle a challenge due to longitudinal and lateral combined vehicle dynamic characteristic and steering uncertainty in the proposed control method. The lateral control accuracy is less than 1.0 m under the particular scenario.