Pranjal Barman
©SHUTTERSTOCK.COM/SHUTTER Z
The demand for electric vehicles (EVs) has increased recently to meet the desire for an ecofriendly transportation system on a global scale. The electrification of vehicles opens up a number of new options and opportunities for doing research in several fields. In a conventional EV, a single electric motor powers the car through reduction gears and the mechanical differential (MD). A crucial component of every gasoline-powered car or internal combustion engine vehicle is the MD, which modifies the speed of the driving wheels during cornering maneuvers. However, the MD has considerable friction losses, is big and heavy, and has a complicated structural design. As a result, the new energy-efficient EV drivetrain uses an electronic differential (ED) in place of the traditional MD. The drivetrains associated with an ED are known as distributed EV drivetrains. Unlike conventional drivetrains, the distributed drivetrains use independently equipped motors to drive the vehicle.
As seen in Fig. 1, the traditional EV drivetrain idea simply employs one electric motor, which is connected to the wheels that are being driven by a reduction gear and an MD system. During a turning maneuver, the MD determines the vehicle speed variances between the inner and outer wheels. A collection of spur gears arranged in various orientations can be used to create the function of the MD. From the open differential to the limited-slip differential, MDs have experienced a lot of modifications. Depending on their intended use, each of these variants has benefits and drawbacks. The main disadvantages of MDs, however, are their complex mechanical design, large size, and heavy weight, which make them inconvenient for EVs, which are designed to be light. Additionally, MDs have high friction loss, which could reduce the driving range of an EV. Therefore, EV designers made certain structural changes to the conventional powertrain, replacing the MD with an ED to lower the curb mass and increase the driving range of EVs.
Fig 1 An EV powertrain with a centralized motor and MD.
This has also been supported by recent advancements in electric motor technology, which have allowed different EV drivetrains to go from pure MDs to EDs. For EVs, Wellington Adams’s wheel-hub–motor drivetrain, often known as an in-wheel motor drivetrain, is one of the innovative drivetrain concepts that is a practical option. This layout eliminates the need for the gearbox, clutch, drive shaft, and MD by directly integrating the motor into the wheels. Despite a considerable reduction in the total bulk and mechanical complexity, this technology makes the tire very heavy, which makes for uncomfortable driving at higher speeds, particularly on uneven ground.
A new hybridization technique that has emerged in recent years is the distributed drivetrain. EVs with a distributed drivetrain architecture are referred to as distributed drive EVs (DDEVs). In a DDEV, the motor and driving wheels are individually connected. The construction of a DDEV is shown in Fig. 2. They can be front-wheel-drive, rear-wheel-drive, or all-wheel-drive vehicles. The benefits of DDEVs include weight reduction, flexibility, controllability, having a quick response, safety, and the opportunity to include cutting-edge control capabilities. The electronic stability control program, advanced driver assistance systems, and acceleration slip regulation are just a few of the advanced control features that are simple to include in DDEVs. The Ford Ecostar powertrain was the first to use the distributed drive layout, and Nissan followed suit with its future electric vehicle concept car.
Fig 2 The DDEV configuration with an ED system.CG: center of gravity.
In DDEVs, the ED is crucial. When the vehicle performs a turning maneuver, the ED is an electronically controlled motor synchronization approach for the separately connected motors with each driving wheel. To achieve stable movement of the vehicle, it distributes enough torque to each wheel so that the speeds of the inner and outer wheels during cornering are precisely synchronized. The ED offers significant advancements over the MD in terms of stability, responsiveness, robustness, and efficiency. Nevertheless, it depends on several sensor feedbacks from the vehicle that require adequate control methodologies to avoid system failures. Therefore, the effective torque distribution method for DDEVs is an emerging research area that creates several possibilities to improve the functionality of EVs.
As was previously mentioned, the ED is a component of a DDEV’s electronic control system. When a four-wheeled vehicle is turning, the inner and outer wheels experience varying rates of rotation. In a DDEV, an ED calculates the required torque speed of the associated motor since the inner wheels have a smaller turning radius than the outer wheels, as illustrated in Fig. 3. Although the ED has many benefits, it also has some shortcomings, including algorithmic failure, noise from the feedback sensors, and correct wheel synchronization in off-road circumstances. The aforementioned problems have been the subject of numerous attempts and in-depth discussions.
Fig 3 The AckermannÐJeantand steering geometry used in an ED system, where L is the wheel base, ${\delta}$ is the turning angle, d is the track width, R is the radius of the turn, VL is the left wheel linear speed, and VR is the right wheel linear speed.
Figure 4 depicts the overall architecture of an ED system. It is an ED drivetrain in which the traction motors are connected to the wheels independently. Each motor must receive speed references from the ED block, which must take data from numerous sensors into account. The Ackermann–Jeantand steering geometry (shown in Fig. 3) is the most popular way to implement the differential action electronically. To fix the wheel mechanism during a turning maneuver, the steering geometry is a structural arrangement of the vehicle steering. Through the use of a closed-loop control technique, Ackermann–Jeantand steering geometry has been employed to accurately determine the required speed and torque for each side wheel. Depending on the turning radius, it can determine the speed of rotation of both the inner and outer wheels.
Fig 4 The architecture of an ED system.
The Ackermann–Jeantand geometry, however, is helpful in quasi-static or low-speed cornering analyses. On a curved path, the radius of the turning center directly relates to the speed of the inner and outer wheels. In this situation, the inner and outer wheel speeds are estimated using the steering angle feedback and the driver’s throttle input. For instance, if the turn is to the right, the ED keeps the left wheel moving at a faster rate than the right wheel. As a function of the steering angle, each driving wheel’s linear speed can be expressed as \begin{align*}{\omega}_{L} & = \frac{{L} + {0.5}\,{\times}\,{d}\,{\times}\,{\tan}\,{\delta}}{L}{\omega}{;} \\ {\omega}_{R} & = \frac{{L}{-}{0.5}\,{\times}\,{d}\,{\times}\,{\tan}\,{\delta}}{L}{\omega} \tag{1} \end{align*} and the speed difference is \[{\Delta}{\omega} = {\omega}_{L}{-}{\omega}_{R} = \frac{{d}\,{\times}\,{\tan}\,{\delta}}{L}{\omega}{.} \tag{2} \]
The direction of the turn can be indicated by the sign of the steering angle; i.e., ${\delta}\,{>}\,{0} = {\text{right}}$, ${\delta}\,{<}\,{0} = {\text{left}}$, and ${\delta} = {0} = {\text{straight ahead}}$. The ED reduces the inner wheel speed and increases the outer wheel speed to balance the vehicle movement. The modified driving speeds by the ED can be given as \[{\omega}_{L}^{\ast} = {\omega} + \frac{\Delta{\omega}}{2}{;}\,{\omega}_{R}^{\ast} = {\omega}{-}\frac{\Delta{\omega}}{2}{.} \tag{3} \]
According to (3), the steering angle is directly proportional to the angular speed of the wheels. Therefore, the ED calculates the required speed of each driving wheel during a cornering maneuvre using steering angle and velocity inputs. Because of its straightforward analysis and straightforward implementation, the Ackermann–Jeantand steering mechanism is used to realize the ED control mechanism in the majority of publications in the existing literature. However, the static steering geometry has some intrinsic flaws, including the inability to account for the impact of centripetal force and the lack of smooth driving control. In addition to the aforementioned tactics, various new sophisticated control methods are currently being suggested to increase the effectiveness and dependability of the steering systems.
A new steering mechanism is found in the motion control and navigation of mobile robots, known as the differential speed steering mechanism. It is simpler than the Ackermann model, having the advantage of achieving a very small turning radius. Active four-wheel-speed-sensitive steering, crab steering, and power steering are the available steering concepts available in steering-related research.
A practical design of an ED can be realized by using microcontroller, sensor, and actuator units. The wheel speed required to accommodate a specific cornering scenario is determined by the microprocessor. It uses the steering angle and velocity sensor feedback and computes the precise speed to be given to the drive motor. Accordingly, it produces the required control signals for the motor controller, which, in turn, manages the drive motors to reach the specified speed. However, this methodology is an open-loop system and, therefore, suffers from stability issues. Closed-loop control continuously monitors the wheel speed and the reference speed and attempts to eliminate errors between them via various control methods, such as proportional-integral-derivative control, sliding mode control (SMC), etc. Most of the published work on EDs uses the steering command, reference torque, and velocity feedback via a differential control algorithm. Steering command-based ED control approaches provide simple, reliable, and stable performance at a limited speed range.
However, some researchers have developed EDs without using the steering commands. Such an ED uses a target slip ratio based on the given speed and road conditions. Then, using a closed control loop, this technique modulates the wheel torque based on reference slip ratios. Another method involves developing a reference speed ratio, calculating the real-time speed ratio based on sensor feedback, and then using that information to generate the necessary torque signal to synchronize the wheel speed throughout cornering of the EV. Apart from the aforementioned architectures, some new ED architectures exist in the literature that are still in the conceptual stage and have not been validated in real-time applications.
The use of ED control mechanisms also differs depending on the different types of electric motors. Modern ED system development can be sped up by using cutting-edge technologies like artificial intelligence, machine learning, the Internet of Things, and blockchain. For forthcoming EVs, there have been numerous opportunities to design ED control systems with improved performance efficiency, cost-effectiveness, robustness, and compactness.
The author would like to acknowledge the support from the Indian Institute of Technology Guwahati–Technology Innovation and Development Foundation and Dr. Anamika Kalita, Department of Science and Technology Inspired Faculty, Institute of Advanced Study in Science and Technology, Guwahati, Assam, India.
• K. Hartani, Y. Miloud, and A. Miloudi, “Electric vehicle stability with rear electronic differential traction,” in Proc. Int. Symp. Environ. Friendly Energies Elect. Appl. (EFEEA), 2010, vol. 58, pp. 1–5.
• S. You, H. Lee, D. Lee, H. Mok, Y. Lee, and S. Han, “Speed ratio control for electronic differentials,” Electron. Lett., vol. 47, no. 16, pp. 933–934, Aug. 2011, doi: 10.1049/el.2011.1546.
• N. Mutoh and T. Takayanagi, “Cornering performance of front and rear wheel independent drive type electric vehicles (FRID EVs) under severe road conditions,” in Proc. 14th Int. IEEE Conf. Intell. Transp. Syst. (ITSC), 2011, pp. 101–106, doi: 10.1109/ITSC.2011.6082965.
• J. F. Daya, P. Sanjeevikumar, F. Blaabjerg, P. W. Wheeler, and J. O. Ojo, “Implementation of wavelet-based robust differential control for electric vehicle application,” IEEE Trans. Power Electron., vol. 30, no. 12, pp. 6510–6513, Dec. 2015, doi: 10.1109/TPEL.2015.2440297.
• A. Haddoun, M. E. H. Benbouzid, D. Diallo, R. Abdessemed, J. Ghouili, and K. Srairi, “Modeling, analysis, and neural network control of an EV electrical differential,” IEEE Trans. Ind. Electron., vol. 55, no. 6, pp. 2286–2294, Jun. 2008, doi: 10.1109/TIE.2008.918392.
• R. N. Jazar, Vehicle Dynamics: Theory and Application. New York, NY, USA: Springer Science Business Media, 2014.
• P. Barman, D. Gogoi, S. Paul, N. Saikia, and S. Sharma, “Steering independent electronic differential based traction control system for independent wheel drive neighborhood electric vehicle,” J. Elect. Syst., vol. 16, no. 4, pp. 498–514, 2020.
Pranjal Barman (pranjalele@gmail.com) earned his M.Tech. and Ph.D. degrees from the Tezpur University, Assam, India. He is a postdoctoral fellow at the Indian Institute of Technology Guwahati–Technology Innovation and Development Foundation, Guwahati Pin-781039, India. His research interests include electric vehicles, power electronics, renewable energy, and underwater exploration.
Digital Object Identifier 10.1109/MPOT.2023.3280338