分布式驱动车辆的动力学控制技术是汽车产业界研究的重点内容。差动转向是该领域的技术之一，它通过使左右轮胎力不同，辅助车辆转向，或促使车辆直接转向。基于轮胎侧偏的差动转向是一种新型转向方式，它充分利用了轮胎侧偏特性，可以减少轮胎的侧向滑移，同时无需转向机构，可降低成本。这种转向方式适用于大曲率半径的转向场景，目前相关研究较少，本文对此种转向方式进行了研究讨论。本文首先建立了分布式驱动车辆的七自由度仿真模型。在此基础上，分析了差动转向的转向过程和系统的稳定性,分析结果表明，车身-轮胎系统是一个稳态系统，施加恒定的差动力矩，可使轮胎保持一定侧偏角，使车辆转向。此外，本文讨论了差动转向的转向能力及其影响因素，结果表明轮胎是影响差动系统转向能力的关键零部件，选用侧偏刚度小的轮胎更容易转向，选用线性区宽的轮胎可以使极限的转向半径更小。一般地，差动转向的车辆极限转向半径为轴距的5倍左右。 本文基于模糊推理前馈补偿的控制方法，设计了一种适用于差动转向车辆的转向控制器。这种方法可以针对不同速度下的转向需求调节前馈补偿量，改善传统比例积分控制的效果。仿真结果表明，该方法可以提升系统的快速性和鲁棒性，使控制器适应不同车速下的转向需求。针对无人小车潜在的轨迹跟踪控制需求，本文进一步讨论了适用于差动转向无人小车的轨迹跟踪算法。算法采用闭环控制的结构，重点考虑了差动转向车辆转向能力有限、执行器易饱和的特点，根据运动学模型和带条件积分的滑模控制算法设计了轨迹跟踪控制器，进行了仿真。仿真结果表明，该控制方法能够良好地完成轨迹跟踪任务，并具有抗积分饱和特性。 本文以集中驱动电动车的传动结构为基础，建立了7自由度的传动系扭振模型，分析了传动系扭振的固有频率和振型，通过实车实验，验证了扭振模型的正确性。以此为基础，本文研究了传动系扭转特性和执行器的延时对差动转向系统转向的影响。仿真结果表明，传动系的扭振特性会使得轮胎纵向力产生波动，但对转向的影响并不显著，而执行器延时能显著降低系统响应的快速性。
The dynamic control technology of distributed drive vehicle has attached increasing attentions in automobile industry. Differential steering is one of the technologies in this field, which can steer the vehicle directly, or assist the vehicle to steer by making the right and left tire forces different. Differential steering based on tire cornering characteristics is a new type of steering. It makes full use of the tire-cornering characteristics, which can reduce the tire side slip, simplify the chassis design and reduce the cost. This new way of steering is applicable to large-radius curve scene. There are few related researches about it, and the paper is devoted to the discussion of it.The paper presents a 7-DOF modeling of a four-wheel independent-driving electric vehicle. According to the dynamic equations of the system, we analyze its steering process and the stability of the system. The results show that it is a steady system. In other word, by applying a constant differential torque the tire can keep a certain sideslip angle and the car will make a turn, after which the steering ability and influencing factors are explored and discussed. The results show that tires are key parts which can affect the steering ability of differential system. Tires with small lateral stiffness helps the vehicle to steer easily, and the tires with a wide linear zone about lateral force characteristic can improve steering capability of steering capability. In general, the minimum steering radius of differential steering vehicles is about 5 times the wheel base.The paper proposes a control method based on fuzzy inference feedforward compensation, which is suitable for steering controller of differential steering vehicles. This method can adjust the feedforward compensation according to the steering demand at different speeds and improve the effect of traditional proportional integral control. Results of the simulation show that the proposed control algorithm not only improves the rapidity of the system response but also can be adapted to different steering needs at different speeds. The paper further discusses the trajectory tracking algorithm for the differential steering unmanned vehicle because of its potential trajectory tracking need. The algorithm adopts the closed-loop structure, which considering the limited steering capability of differential steering system and actuator saturation problem. A trajectory tracking controller is designed by using kinematic model and sliding mode control algorithm with conditional integrators. Results of the simulation show that the controller performs well and has the anti-integral saturation characteristic.Finally, based on the drivetrain structure of centrally-driven electric vehicle, a torsional vibration model with 7 degrees of freedom is established. According to this model, the paper analyzes the natural frequency and mode of vibration of the drivetrain and then designs the experiment to verify it. After on that, the paper goes on to study the effect result from the torsional vibration characteristics of the driveline and the delay of the actuator. Results shows that the torsional vibration characteristics of the driveline usually leads to fluctuation about the tires longitudinal force, but make unobvious effect on steering. Actuator delay can reduce the rapidity of system response significantly.
