New Control Strategy Boosts Electric Vehicle Stability
The automotive industry is currently witnessing a profound transformation driven by electrification and intelligence. As traditional internal combustion engine vehicles give way to electric architectures, the fundamental design of the vehicle chassis is undergoing a radical reimagining. Among the most promising developments is the distributed drive electric vehicle. Unlike conventional cars that rely on a central motor and mechanical transmission systems to deliver power, distributed drive vehicles equip each wheel with its own independent electric motor. This configuration offers unprecedented control over torque distribution, promising enhanced efficiency and performance. However, this complexity introduces significant challenges in maintaining vehicle stability, particularly during extreme maneuvers or on low-friction surfaces.
Addressing these critical safety concerns, a research team from the School of Automobile and Transportation Engineering at Liaoning University of Technology has developed a sophisticated collaborative control strategy. Their work focuses on improving the lateral stability of distributed drive electric vehicles by integrating active front steering with direct yaw moment control. The findings, recently published, offer a robust solution to the coupling effects between steering and driving systems, marking a significant step forward in chassis electrification technology.
The Challenge of Distributed Drive Architecture
The shift toward distributed drive systems represents a leap in automotive engineering. By eliminating heavy mechanical transmission components and installing drive motors directly within or near the wheels, manufacturers can achieve better space utilization and energy efficiency. More importantly, the independent controllability of the four wheels allows for precise torque vectoring. This capability can generate additional yaw moments to correct vehicle attitude, theoretically offering superior handling compared to traditional stability control systems.
However, the path to realizing this potential is fraught with technical hurdles. The primary issue lies in the strong coupling between the steering system and the drive or braking systems. In a distributed drive vehicle, changes in driving torque affect the tire forces, which in turn influence the steering behavior. Conversely, steering inputs alter the load distribution on the wheels, impacting the available traction for drive motors. If these systems are controlled independently, they may work at cross-purposes. For instance, the steering system might attempt to correct a slide while the torque distribution system inadvertently exacerbates it. Therefore, a coordinated approach is not just beneficial but necessary to ensure safety and performance.
Previous research has explored various methods to manage this complexity. Some studies have utilized adaptive control algorithms to integrate multi-input and multi-output models, while others have employed fuzzy logic or game theory to coordinate steering and driving stability. Despite these advancements, there remains a need for control strategies that are both highly adaptive to changing road conditions and robust against system uncertainties. The research from Liaoning University of Technology addresses this gap by proposing a layered control architecture that dynamically balances the intervention of steering and torque systems.
A Layered Control Architecture
The core innovation proposed by the researchers is a two-layer control strategy designed to harmonize the Active Front Steering system and the Direct Yaw Moment Control system. This hierarchical approach separates the decision-making process from the execution, allowing for more precise management of vehicle dynamics.
The upper layer serves as the collaborative controller. Its primary function is to assess the real-time stability of the vehicle and determine how much control authority should be assigned to the steering system versus the torque distribution system. To make this determination, the researchers utilized a phase plane strategy based on the vehicle’s centroid sideslip angle and sideslip angle velocity. In vehicle dynamics, the sideslip angle represents the difference between the direction the vehicle is pointing and the direction it is actually traveling. The rate at which this angle changes is a critical indicator of instability.
By mapping these parameters on a phase plane, the researchers defined specific stability regions. The boundaries of these regions were established using a double-line method, which accounts for varying road adhesion coefficients. This allows the controller to recognize when the vehicle is operating within a safe zone, when it is approaching the limits of stability, and when it has entered a critical state requiring immediate intervention. Based on the vehicle’s position relative to these boundaries, the upper controller calculates coordination coefficients. These coefficients act as weights, dictating whether the Active Front Steering system, the Direct Yaw Moment Control system, or both should be active.
For example, when the vehicle is operating well within the stable region, the system may rely primarily on steering adjustments, which are generally smoother and less intrusive for the driver. As the vehicle approaches the stability boundary, such as during a high-speed lane change on a wet road, the coordination strategy shifts. Both systems are engaged to work in tandem. If the vehicle exceeds the collaborative control region and enters a critical instability zone, the system prioritizes direct yaw moment control, as generating torque differences between wheels can be more effective than steering alone in recovering from a spin or severe slide. This dynamic allocation ensures that the control systems are used optimally without unnecessary intervention that could disturb the driver.
Robust Execution via Sliding Mode Control
Once the upper controller determines the required level of intervention, the lower layer controllers execute the specific commands. The researchers employed variable structure sliding mode control for this execution layer. Sliding mode control is renowned in engineering for its robustness against uncertainties and disturbances. In the context of a vehicle, uncertainties can arise from changes in vehicle load, tire wear, or unpredictable variations in road surface friction.
The Active Front Steering controller in the lower layer calculates the necessary additional front wheel angle to track the desired yaw rate. It uses a sliding surface defined by the error between the actual and expected yaw rates. To prevent the chattering effect often associated with sliding mode control, which can cause mechanical wear and discomfort, the researchers incorporated a saturation function. This smooths the control input while maintaining the system’s ability to converge quickly to the desired state.
Simultaneously, the Direct Yaw Moment Control controller calculates the additional yaw moment required to stabilize the vehicle. This controller considers both the yaw rate error and the sideslip angle error. The weighting between these two errors is not fixed; it adapts based on the road adhesion coefficient. On low-friction surfaces, such as ice or wet asphalt, the controller places greater emphasis on controlling the sideslip angle to prevent loss of traction. The calculated yaw moment is then distributed among the four wheels.
The torque distribution logic takes into account the vertical load on each axle. During acceleration or braking, weight transfers between the front and rear axles, altering the maximum traction each wheel can provide. The control algorithm adjusts the torque sent to each motor based on these load ratios, ensuring that no wheel is asked to produce more force than the tire-road interface can support. This prevents wheel spin and maximizes the effectiveness of the stability correction. The individual motor dynamics are also modeled, accounting for the response time of the electric motors to ensure the commanded torque is delivered accurately and swiftly.
Rigorous Validation Through Hardware-in-the-Loop Testing
To verify the effectiveness of this collaborative control strategy, the research team conducted extensive Hardware-in-the-Loop tests. This testing method is considered a gold standard in automotive development because it bridges the gap between pure simulation and real-world road testing. In a Hardware-in-the-Loop setup, the physical controller hardware is connected to a real-time simulation of the vehicle and its environment. This allows engineers to test dangerous scenarios, such as loss of control on ice, without risking damage to a prototype vehicle or injury to a driver.
The test bench utilized in this study was constructed using a combination of industry-standard software and hardware. The vehicle dynamics and simulation scenarios were modeled using CarsimRT, which provides high-fidelity representations of vehicle behavior. The controller models were built and compiled using Matlab and Simulink, then deployed to a dSPACE controller. This controller communicated with the hardware platform via a CAN bus, mimicking the network architecture found in modern production vehicles. A driving simulator allowed human operators to input steering and pedal commands, adding a layer of realism to the test conditions. Sensors captured vehicle signals in real-time, feeding them back to the controller for closed-loop operation.
The researchers selected two specific test scenarios to evaluate the system: a double lane change maneuver and a step steer input. The double lane change test simulates an emergency avoidance situation where a driver must quickly swerve to avoid an obstacle and then return to the original lane. This maneuver places high demands on lateral stability and is a standard procedure for evaluating vehicle handling. The tests were conducted on a simulated wet road surface with a low adhesion coefficient of 0.45, representing challenging winter or rainy conditions. The vehicle speed was set to 80 kilometers per hour, a speed at which stability control systems are critical.
The second scenario involved a step steer input, where the steering wheel is turned abruptly to a fixed angle. This test evaluates the vehicle’s transient response and its ability to settle into a stable state after a sudden disturbance. For this test, the steering input was set to 90 degrees, again on a low-friction surface at 80 kilometers per hour. These conditions were chosen to push the vehicle to the limits of its stability envelope, providing a rigorous stress test for the control algorithms.
Performance Results and Analysis
The data collected from the Hardware-in-the-Loop tests provided compelling evidence of the collaborative control strategy’s superiority. When compared to scenarios where the Active Front Steering and Direct Yaw Moment Control systems operated independently, the collaborative approach demonstrated significant improvements in key stability metrics.
During the double lane change maneuver, the vehicle equipped with the collaborative controller exhibited much smoother response curves for both yaw rate and sideslip angle. The peak values of these parameters were notably lower than those observed in the independent control cases. Specifically, the peak yaw rate was reduced by 10 percent, and the peak sideslip angle was reduced by 14 percent. These reductions are substantial in the context of vehicle safety. A lower sideslip angle means the vehicle is less likely to slide sideways out of the driver’s intended path, while a controlled yaw rate ensures the vehicle rotates predictably around its vertical axis. The results indicate that the collaborative system effectively suppressed skidding phenomena on the low-friction surface, allowing the vehicle to track the desired path with greater accuracy.
In the step steer test, the benefits of the collaborative strategy were equally evident. The vehicle reached a stable state more quickly and with less oscillation. The response curves showed lower overshoot and fewer vibrations before settling. Quantitatively, the peak yaw rate and sideslip angle were reduced by 6 percent and 13 percent, respectively, compared to independent control. Perhaps more importantly, the time required to reach a steady state was shortened by 2 seconds. In an emergency situation, those two seconds can be the difference between a successful avoidance maneuver and a collision. The faster stabilization implies that the driver regains control of the vehicle sooner, reducing panic and improving overall safety margins.
The study also highlighted the effectiveness of the phase plane-based upper controller. By accurately identifying the stability regions, the system was able to intervene only when necessary and with the appropriate intensity. This prevents the over-activation of stability systems, which can sometimes degrade the driving experience or reduce energy efficiency. The seamless coordination between steering and torque distribution ensured that the vehicle behaved predictably, maintaining the driver’s confidence even in extreme conditions.
Implications for the Future of Automotive Safety
The findings from Liaoning University of Technology have broad implications for the future of automotive safety and autonomous driving. As vehicles become increasingly electrified, the reliance on software-based control systems to manage dynamics will only grow. Traditional mechanical limits are being replaced by electronic boundaries defined by control algorithms. The collaborative control strategy demonstrated in this research provides a blueprint for how these systems can be integrated to maximize safety without compromising performance.
For manufacturers of distributed drive electric vehicles, this research offers a viable path to overcoming the stability challenges inherent in the technology. By adopting a layered control architecture that utilizes phase plane analysis and sliding mode control, engineers can develop chassis systems that are robust against varying road conditions and driving styles. This is particularly relevant for high-performance electric vehicles, where the instant torque delivery of electric motors can easily overwhelm tire traction if not managed correctly.
Furthermore, this technology is a critical enabler for autonomous driving. Self-driving vehicles must be able to handle emergency maneuvers safely without human intervention. A stability control system that can coordinate steering and braking independently and effectively is essential for the safety certification of autonomous platforms. The ability to maintain stability on wet or slippery roads expands the operational design domain of autonomous vehicles, allowing them to function safely in a wider range of weather conditions.
The use of Hardware-in-the-Loop testing in this study also underscores the importance of rigorous validation in the development of safety-critical systems. As control algorithms become more complex, simulation and real-time testing become indispensable tools for ensuring reliability. The methodology employed by the research team can serve as a standard for future developments in chassis control, ensuring that new features are thoroughly vetted before reaching the road.
Conclusion
The transition to distributed drive electric vehicles represents a significant opportunity to enhance vehicle performance and efficiency. However, realizing this potential requires advanced control strategies to manage the complex interactions between steering and drive systems. The research conducted by Wang Cheng, Qu Xiaozhen, and Sun Xiaobang at Liaoning University of Technology provides a significant contribution to this field. Their collaborative control strategy, which integrates Active Front Steering and Direct Yaw Moment Control through a phase plane-based upper controller and sliding mode lower controllers, has been proven to significantly improve lateral stability.
Through rigorous Hardware-in-the-Loop testing under challenging low-friction conditions, the study demonstrated that collaborative control outperforms independent system operation. The reductions in yaw rate and sideslip angle peaks, along with faster stabilization times, confirm that this approach can effectively prevent loss of control during emergency maneuvers. As the automotive industry continues to evolve toward electrification and automation, such innovations in chassis control will be fundamental to ensuring the safety and reliability of next-generation vehicles. The work stands as a testament to the importance of academic research in solving practical engineering challenges, bridging the gap between theoretical control dynamics and real-world vehicle safety.
Authors: Wang Cheng, Qu Xiaozhen, Sun Xiaobang Affiliation: School of Automobile and Transportation Engineering, Liaoning University of Technology, Jinzhou 121000, China Journal: Modern Manufacturing Engineering DOI: 10.16731/j.cnki.1671-3133.2023.11.009 ISSN: 1671-3133 Publication Year: 2023 Issue: 11 Pages: 63-70