Research Projects

Clockwise from upper left: Zhichao Shen, Mingkang Wu, Kevin Li, Albert Chen, Ao Yu, James Van Wagnen, Handan Shu, Luna Zhang, Prof. Chin-An Tan, Prof. Buyun Zhang
Not pictured: Shahram Amoozegar, Minjun Bae, Grace Sexton, Sai Vijay Gorle, Justin Hadyniak

Path Planning for Automated Valet Parking System

Mingkang Wu, visiting scholar from Jiangsu University
Kevin Li, Alumnus
Sai Vijay Gorle, Undergraduate Student

For an automated parking system, path planning focuses much on the parking process. RRT (Rapid-exporting Random Trees) can be used in this case. The advantage of the RRT algorithm is fast speed. During the auto-parking process, every time the vehicle moves, the path planning algorithm will re-generate a path for the vehicle to follow.

Based on the automated parking system, the Automated Valet Parking System is introduced, which can effectively improve people's travel efficiency, save the time needed to find parking spots, and reduce the number of accidents caused by unskilled driving. A simplified model of a parking lot and vehicle is drawn in the software, which is convenient for the testing algorithm. Through programming in the software to show the various parameters of the vehicle, including length, width, and steering angle, it is convenient to observe the state of the vehicle directly.

Smart Vehicle Suspension

Handan Shu, Graduate Student
Zhichao Shen, Graduate Student
Professor Buyun Zhang, visiting scholar from Jiangsu University

Schematic of Electronically Controlled Air Suspension (ECAS)

Nowadays, air suspension systems are widely applied on ground vehicles, but there are still some problems existing in its implementation such as vehicle stability, reliability of system, ride comfort, etc. Our group is developing a reliable electronically controlled air suspension (ECAS), which can achieve vehicle stability in roll and yaw movement as well as maintain ride comfort by decreasing the acceleration in the vertical and pitch directions. A robust algorithm is also explored to deal with all the requirements and calculate an optimized result. The program is currently in progress and we are making efforts to develop a strong control system and test it on the vehicle after acquiring a good simulation result.

Intelligent Suspension Systems of Autonomous Vehicles with Incorporation of Road Preview

Grace Sexton, Undergraduate Student
Handan Shu, Graduate Student
Zhichao Shen, Graduate Student
Professor Buyun Zhang, visiting scholar from Jiangsu University

It is estimated that in the United States alone, several billion dollars in vehicle damage are caused by potholes on the road. Although most Advanced Driver-Assistance Systems (ADAS) have implemented the use of controls, sensors, and thorough algorithms for the control of vehicle suspension systems, there is still a lack of road preview techniques being incorporated into such control systems—which creates the potential for communication discrepancies between the vehicle and its surrounding terrain. By adding image processing methodologies to the suspension control system, this environmental layer will reduce the time it takes for the vehicle to identify a road disturbance and subsequently adjust the suspension system accordingly. This will allow for a safer and more reliable drive while also preserving the resilience of the suspension system’s components.

In order to improve vehicle suspension with the use of road preview, it is necessary to first assess the current methods of suspension control, as well as any recent advancements within the realms of image processing techniques. Then, through extensive dynamic modeling and simulation, the best method of road preview integration will be determined based on several factors, including the amount of time it takes for the vehicle to identify and adjust to an object of disturbance, the quality of vibration control that ensues, and the costs associated with such integration.

ADAS for Safe Curve Maneuvering

Luna Zhang, Alumnus
Handan Shu, Graduate Student
Zhichao Shen, Graduate Student

Advanced driver assistance systems (ADAS) such as adaptive cruise control and lane centering keeping can greatly reduce the driver’s burden, especially during highway and extended driving. The human driver adjusts the steering angle to keep the vehicle in the center of lane through looking ahead of the road and sensing the changes in vehicle position. In ADAS, a camera takes this responsibility of detecting lanes, analyzing lane information and estimating the relative position between the lanes of both sides and ego vehicle in the vehicle coordinates. The information is delivered to a controller, which decides a steering angle executed by the vehicle. This problem is more challenging in curvy maneuvering because the vehicle cannot be seen as a rigid body and there may exist lateral slip (not in steady state) and other physical limitations, as well as a discontinuous curvature change of road profile, which increases the difficulty of control.

Autonomous Robotic Lawnmower

James Van Wagnen, Graduate Student
Ao Yu, Alumnus

Current robotic lawnmowers in industry face several problems that have impeded their ability to overtake push mowers in the market. This is mainly due to their expensive cost and added operating requirement of a boundary wire. We are currently exploring the innovative utilization of simultaneous localization and mapping (SLAM) in the algorithm of a robot lawnmower to remove the boundary wire requirement, optimizing it for future commercialization. This is done using a 3D camera and a commercially available programmable robot. The project is still incomplete; however, further work will aim to validate the SLAM algorithm as well as the whole robot.

Collaborative Robot Research

Albert Chen, Graduate Student
Justin Hadyniak, Undergraduate Student

Traditional robots have been widely used in manufacturing industries for many years; however, in everyday life, the applications of robots are still limited. There are several reasons:
1. Safety
2. Tradition robots lack flexibility and adaptability
3. Cost
Collaborative robots were designed to tackle those problems. Integration of co-robots to assist humans in every aspect of life can lead to a promising future for human beings.

Optimization of Constrained, Distributed Piezoelectric Vibration Energy Harvesting Systems

Shahram Amoozegar, Graduate Student

Over the past two decades, many innovations, particularly in the area of MEMS research, have made it possible to design micro sensors and actuators that are capable of operating with very low power requirements, thus allowing sensor networks to be deployed and operated in an autonomous manner. Some of these sensors are used in applications with self-contained power sources packaged with the sensors. While batteries are currently still used in many of these applications, piezoelectric vibration energy harvesters have shown a lot of potential for use with self-powered devices in applications where ambient vibration energy is present. The development of self-powered sensors and devices not only reduces environmental hazards related to battery disposals, improves battery lifetime, and in some cases, alleviates the burden of carrying bulky batteries, it also presents new opportunities and challenges in the emerging research field of renewable energy sources for sensors and electronic systems.

In this project, a systematic approach based on distributed transfer functions is employed to investigate the response and optimal design of piezoelectric vibration energy harvesters under arbitrary boundary and constraint conditions. A computer code is developed to allow parameter optimization of these harvesters. As an example, the analysis and computer code can be easily applied to understand the optimized harvested voltage for a constrained beam energy harvester as shown below.

Figure: Cantilevered piezoelectric unimorph energy harvester with an intermediate rotational spring.

Novel-Structured Nanomaterials for Next Generation Energy Storages, and Simulation of Microstructural Evolution and Charge Distribution During Lithium Electrodeposition

Minjun Bae, Graduate Student (co-advised)
Professor Da Deng (co-advisor), Chemical Engineering Department

Although lithium-ion (Li-ion) batteries have been attracting much attention as a power source for electronic devices, such as laptops and cell phones, conventionally used graphitic anodes have almost reached to its maximum capacity (372 mAh/g), which is not sufficient to satisfy the increasing demand of higher-energy-density batteries for future energy storage applications. The electrochemical performance of a battery cell is mainly determined by electrochemical reactions between two electrode materials. Therefore, researchers have been motivated to explore alternative electrode materials to improve cyclability, energy density, capacity and Coulombic efficiency of a battery cell. Among various candidates, lithium (Li) metal is considered as one of the most promising alternative anode materials due to its superior specific capacity (3860 mAh/g), and the lowest electrochemical potential (-3.040 V versus standard hydrogen electrode). However, the intrinsic Li dendrite growth during charging cycle causes severe safety concerns and poor cycling behavior. Herein, we have developed a novel titanium current collector to efficiently avoid the undesirable Li dendrite formation (Figure 1).

The objective of this simulation project is to gain a deep understanding of underlying physics of Li dendrite growth. Experimental results showed that the Ti current collector significantly improved cyclability and Coulombic efficiency. Theoretical simulation of Li electrodeposition behavior using FEA method is also under investigation (Figure 2).