面向复杂场景的AWID/AWIS无人车动力学智能集成控制方法研究

The Unmanned Ground Vehicle (UGV) has great potentials to replace humans to conduct various missions. Therefore, it is expected to be widely used in both military and civilian applications. Each wheel of the All-wheel-independently-driven/All-wheel-independently-steered (AWID/AWIS) UGV can be driven or steered independently to improve the flexibility, mobility, and handling stability. Therefore, The AWID/AWIS UGV becomes very hot in the field of the research of high-performance military or civilian UGV. However, the application of the AWID/AWIS UGV faces several great challenges. First, the autonomous working condition is always very complex for AWID/AWIS UGV. Second, the uncertainties and external disturbances of the system are quite large. Third, the Degree of Freedoms (DOFs) of the controllable actuators of the AWID/AWIS UGV are much higher than those of the conventional vehicle. Fourth, the coordinated dynamics control of the chassis is quite difficult for the AWID/AWIS UGV. To overcome above problems, this project aims at proposing the intelligent integrated dynamics control approach of the AWID/AWIS UGV. The uncertainties and the external disturbances of the dynamics system and the planning path of the AWID/AWIS UGV will be revealed, as well as the multi-scale delay characteristics of the complex Networked Control System (NCS). Based on this, the closed-looped vehicle dynamics system of the AWID/AWIS UGV considering complex system uncertainties could be built to understand the accurate dynamics behavior of the UGV in the typical motion. In addition, the reconfiguration approach of the steering mode of the AWID/AWIS UGV to meet the requirement of the complex working condition and complex path will be proposed. Based on the double-envelope method that describes the driving limit of the AWID/AWIS UGV, the intelligent integrated dynamics control approach for the AWID/AWIS UGV, which satisfies the complex autonomous driving control target and has the ability to reconfigure the steering mode to operate in the complex different environments, will be proposed. In addition, the intelligent integrated dynamics control approach is supposed to handle the nonlinear coupling characteristics of the vehicle dynamics induced by all the AWID/AWIS wheels. The expected research results and contributions will improve the flexibility, mobility, and handling stability of the AWID/AWIS UGV, as well as providing great scientific significance for the development of the high performance military and civilian UGV.

无人车可替代人类执行各类任务，在军用及民用领域有广泛应用前景。AWID/AWIS无人车采用各车轮独立驱动及独立转向技术，是高性能军用及民用无人车的研究热点。本项目针对AWID/AWIS无人车自主驾驶场景复杂、系统不确定性大、可控执行机构自由度高、动力学协调控制难等问题，开展其动力学智能集成控制方法研究，具体为：揭示整车动力学系统不确定性、自主驾驶路径信息不确定性、复杂网络化控制系统多尺度时滞等扰动的产生机制，构建包含多维不确定性的无人车闭环动力学模型，掌握典型工况动力学特性，探索面向复杂多变路径的多转向模式自主切换、组合及重构机制，结合驾驶极限双包络线描述方法，构建面向自主驾驶多控制目标、对复杂环境有自适应重构能力、自主协调各车轮独立作动单元耦合特性的动力学智能集成控制方法，提高AWID/AWIS无人车的灵活性、机动性、操纵性及稳定性。研究成果将为高性能军用及民用无人车研制提供理论支撑。

可重构车辆的系统设计发生革命变化，基本构型元素为可重构运载单元，以实现“组小成大、化整为零”。可重构运载单元具备自主行驶、自主组合、自主解体能力以及机械系统（物质流）—能量系统（能量流）—控制系统（信息流）可重构特征。提出以单桥双轮构型可重构运载单元作为基本重构单元的可重构车辆基本构型，形成可重构运载单元“形态重构”“控制重构”“能量重构”三个维度重构理念，突破可重构运载单元间物质流-能量流-信息流的融合与解体技术，分析由于模块化单元动态重构所造成的多维参数不确定性，建立包含参数不确定性和路径信息的车辆路径跟踪闭环动力学模型。提出“随控布局”设计可重构车辆的高敏捷操纵控制方法，提高可重构车辆总体布局和动态重构的灵活性。分析可重构车辆的目标闭环极点位置描述方法，构建直接矢量控制架构及质心目标运动矢量跟随鲁棒控制律，实现全轮独立驱动/制动/转向可重构车辆“能力到边”的高机动性控制，采用“G-G”图描述高速可重构无人车极限动力学性能，提出面向极限动力学性能的纵向-横向轨迹跟踪控制器，极大地提升高速可重构无人车的轨迹跟踪控制性能和稳定性能。采用云端大数据技术实现可重构运载单元的云端-边缘端群体动力学控制，提出云端-边缘端联合控制架构和虚拟领航者的云-车双向队列架构，在云端模块提出基于机器学习方法的单元控制器参数优化机制，建立包含通讯不确定性的群体可重构单元闭环动力学模型，实现群体可重构运载单元的一致性控制和鲁棒性分析。