- Smart Agriculture and AI
- Soil Mechanics and Vehicle Dynamics
- Agricultural Engineering and Mechanization
- Remote Sensing in Agriculture
- Soil Management and Crop Yield
- Leaf Properties and Growth Measurement
- Advanced Combustion Engine Technologies
- QR Code Applications and Technologies
- Energy, Environment, and Transportation Policies
- Electric and Hybrid Vehicle Technologies
- Stability and Control of Uncertain Systems
- Vehicle emissions and performance
- Photovoltaic Systems and Sustainability
- Electric Vehicles and Infrastructure
- Crop Yield and Soil Fertility
- Advanced Algorithms and Applications
- Image Processing Techniques and Applications
- Biodiesel Production and Applications
- Photovoltaic System Optimization Techniques
- Robotic Path Planning Algorithms
- solar cell performance optimization
- Adaptive Control of Nonlinear Systems
- Advanced Battery Technologies Research
- Plant Surface Properties and Treatments
- Advanced Materials and Mechanics
Nanjing Agricultural University
2014-2025
University of Illinois Urbana-Champaign
2010
Clemson University
2009
CAE (Canada)
2007
It is necessary to detect multi-type farmland obstacles in real time and accurately for unmanned agricultural vehicles. An improved YOLOv5s algorithm based on the K-Means clustering CIoU Loss function was proposed improve detection precision speed up real-time detection. The used order generate anchor box scales accelerate convergence of model training. function, combining three geometric measures overlap area, center distance aspect ratio, adopted reduce occurrence missed false precision....
In the field of agricultural robotics, robotic grippers play an indispensable role, directly influencing rate fruit damage and handling efficiency. Currently, traditional face challenges such as high rates requirements for position control. A gripper stable spherical with positional offset tolerance a low is proposed in this paper. It adopts three-finger structure. flexible active roller configured at end each finger, allowing translation just gentle touch. An integrated pressure sensor...
A vision-based row guidance method is presented to guide a robot platform which designed independently drive through the crops in field according design concept of open architecture. Then, offset and heading angle are detected real time on basis recognition crop using machine vision. And control scheme proposed carry out guidance. Finally, preliminary experiments were implemented vegetable field. Experimental results show that algorithms identification effective parameters measured analyzed...
For output-constrained switched systems, uniformly local stabilization problem via state feedback is addressed. A unified tool, a common tangent-type barrier Lyapunov function (Tan-BLF), developed to deal with the of systems and cope constraints imposed on system output or without. Then, combining adding power integrator technique (APIT) function, state-feedback stabilizing laws are designed systematically make resultant closed-loop asymptotically stable while preventing violation...
Photovoltaic energy has shown a drastic increase in recent years, and photovoltaic greenhouses, as new modes of distributed power generation combined with agricultural can yield profit from besides planting income, while easing the pressure on supply land resources for station promoting application products agriculture, compared to traditional greenhouses. Research economic benefits are not enough global manner promote Therefore, models costs income greenhouses were established, an...
In order to remedy the defects of single sensor in robustness, accuracy, and redundancy target detection, this paper proposed a method for detecting obstacles farmland based on information fusion millimeter wave (mmWave) radar camera. Combining advantages mmWave range speed measurement camera type identification lateral localization, decision-level algorithm was designed information, global nearest neighbor used data association. Then, effective sequences with successful association were...
Although spraying pesticides onto stems and leaves is the most effective method to prevent plant pests diseases, splashing of droplets upon impact with leaf surface one main ways that liquid agent lost during application process. Moreover, collision flowing film are complex owing presence microstructures on surface. Therefore, understanding droplet dynamics crucial for improving pesticide In this paper, a simulated rice microstructured was established, impacting thin were investigated using...
A control method of neural network controller with reinforcement learning is proposed to implement idle speed an automobile engine reduce fluctuation the speed. Firstly, reinforcement-learning demonstrated in detail. Then, scheme designed experiment. Q algorithm, as one methods learning, used for network, which based on evaluating system performance and giving credit successful actions. After trained fully, contrast experiments are implemented test bench between original controller....
A new method of vision-based guidance line detection is developed to find in row crop fields for agricultural machinery, which uses the mathematical morphology process images and segment crops extract target crops. The fit with least-square according centroids individual or/and central points several pixel rows a continuous region image. could be found easily three kinds fields, such as plants spacing image, both features above. Then experiments were implemented by an robot vegetable field,...
Solar-powered agricultural vehicles facilitate the reduction in use of fossil fuel, decreasing emissions air pollutants and greenhouse gases. This paper assessed electric vehicle (AEV) based on photovoltaics China, taking into consideration three aspects, i.e., power supply mode for AEV, economy analysis. The mode, whether on-board photovoltaic (PV) system or off-board PV system, is subjected to size designed but AEV's depends area panels which affects initial investment cost AEV. Three...
It is difficult to establish the precise mathematical model of agricultural wheeled robots with differential drive for path tracking control, due characteristics nonlinear, strong coupling and multivariable. Here, control studied robot based on sliding mode variable structure. Firstly, motion established goal determined tracking. Then, structure applied design controller. Finally, simulations were conducted circle sine curves verify controller’s performance, further real tests carried out a...
Tree trunks detection and their location information are needed to perform effective production management in forestry fruit farming. A novel algorithm based on data fusion with a vision camera 2D laser scanner was developed detect tree accurately. The transformation built from coordinate system an image system, the model of rectangle calibration plate two inward concave regions established implement alignment between sensors data. Then, decision Dempster-Shafer theory were achieved through...
This article discusses the development of variable field view (FOV) camera to realize headland turning an agricultural robot in corn fields. The FOV was implemented change direction by two DC motors rotating separately vertical and horizontal planes. Headland is executed six steps: end row detection guidance, going blind for a distance, first 90˚ turning, position calculation, backing control, second turning. Mathematically morphological operations were chosen segment crops, fuzzy logic...
A multipurpose robot was designed to perform several tasks such as spraying and weeding in a greenhouse. Several operations were conducted by adding or removing sensing component(s), replacing actuator(s) switching control software, with little no change of the platform. sliding mode applied motion light its kinematic model. Vision based algorithms developed for navigation row planting crops, taking water machine examples. vertical projection method calculate guidance line navigation....