Jingyao Gai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jingyao Gai has a total of 51 co-authors. Here are the top co-authors by number of shared publications:
- Zhiyang Guo
School of Mechanical Engineering, Guangxi University, Nanning 530004, China
3 shared publications - Chuanrong Huang
College of Mechanical Engineering, Guangxi University, Nanning, China
2 shared publications Lie Tang
2 shared publications- Lie Tang
Iowa State University
2 shared publications Lirong Xiang
2 shared publicationsMingchang Zhang
2 shared publications
Latest Publications
- Open Access
A Safe and Efficient Navigation Framework for Ground Vehicles on Uneven Terrain Considering Kinematic Constraints and Terrain Traversability
Ground vehicles navigating uneven terrain must simultaneously guarantee motion safety and efficiency. Safety requires that the planned waypoints lie in highly traversable terrain, while ensuring vehicle reachability to these waypoints, which must be kinematically feasible. Efficiency demands fewer detours and smoother paths that avoid excessive vehicle acceleration and steering. However, existing path planning research for uneven terrain fails to comprehensively integrate vehicle kinematic constraints, terrain factors, path smoothness, rollover risk, and total path length. To address this problem, this paper proposes a novel navigation framework. It first integrates terrain slope, flatness, elevation variation, and sparsity to generate a 2D global terrain traversability cost map. Subsequently, a three-phase path planning algorithm integrates A*, guided Rapidly-exploring Random Tree (RRT), and our proposed Kinematic and Terrain-Aware Probabilistic Roadmap (KT-PRM) local re-planning algorithm, which jointly considers multiple factors including ground vehicle kinematic constraints, terrain factors, path smoothness, rollover risk, and path length. This three-phase combination delivers safe, smooth, and short global paths over uneven terrain within a relatively short planning time. Finally, Nonlinear Model Predictive Control (NMPC) is employed for path tracking in the framework. Experiments were conducted in both simulated and real-world uneven terrain environments. The results demonstrated that the three-phase path planning algorithm integrated with our proposed KT-PRM algorithm achieves comprehensive performance in generating safer, smoother, and shorter paths. Our proposed navigation framework achieves safer and more efficient navigation compared with existing navigation frameworks.
Open AccessA Safe and Efficient Navigation Framework for Ground Vehicles on Uneven Terrain Considering Kinematic Constraints and Terrain Traversability
Ground vehicles navigating uneven terrain must simultaneously guarantee motion safety and efficiency. Safety requires that the planned waypoints lie in highly traversable terrain, while ensuring vehicle reachability to these waypoints, which must be kinematically feasible. Efficiency demands fewer detours and smoother paths that avoid excessive vehicle acceleration and steering. However, existing path planning research for uneven terrain fails to comprehensively integrate vehicle kinematic constraints, terrain factors, path smoothness, rollover risk, and total path length. To address this problem, this paper proposes a novel navigation framework. It first integrates terrain slope, flatness, elevation variation, and sparsity to generate a 2D global terrain traversability cost map. Subsequently, a three-phase path planning algorithm integrates A*, guided Rapidly-exploring Random Tree (RRT), and our proposed Kinematic and Terrain-Aware Probabilistic Roadmap (KT-PRM) local re-planning algorithm, which jointly considers multiple factors including ground vehicle kinematic constraints, terrain factors, path smoothness, rollover risk, and path length. This three-phase combination delivers safe, smooth, and short global paths over uneven terrain within a relatively short planning time. Finally, Nonlinear Model Predictive Control (NMPC) is employed for path tracking in the framework. Experiments were conducted in both simulated and real-world uneven terrain environments. The results demonstrated that the three-phase path planning algorithm integrated with our proposed KT-PRM algorithm achieves comprehensive performance in generating safer, smoother, and shorter paths. Our proposed navigation framework achieves safer and more efficient navigation compared with existing navigation frameworks.
- Open Access
Estimation of Chlorophyll and Water Content in Maize Leaves Under Drought Stress Based on VIS/NIR Spectroscopy
Maize (Zea mays) is a key crop, with its growth impacted by drought stress. Accurate, non-destructive assessment of drought severity is crucial for precision agriculture. VIS/NIR reflectance spectroscopy is widely used for estimating plant parameters and detecting stress. However, the relationship between key parameters—such as chlorophyll and water content—and VIS/NIR spectra under drought conditions in maize remains unclear, lacking comprehensive models and validation. This study aims to develop a non-destructive and accurate method for predicting chlorophyll and water content in maize leaves under drought stress using VIS/NIR spectroscopy. Specifically, maize leaf reflectance spectra were collected under varying drought stress conditions, and the effects of different spectral preprocessing methods, dimensionality reduction techniques, and machine learning algorithms were evaluated. An optimal data processing pipeline was systematically established and deployed on an edge computing unit to enable rapid, non-destructive prediction of chlorophyll and water content in maize leaves. The experimental results demonstrated that the combination of stepwise regression (SR) for feature selection and a stacking regression model achieved the best performance for chlorophyll content prediction ( = 0.8740, RMSEp = 0.2768). For leaf water content prediction, random forest (RF) feature selection combined with a stacking model yielded the highest accuracy ( = 0.7626, RMSEp = 4.12%). This study confirms the effectiveness and potential of integrating VIS/NIR spectroscopy with machine learning algorithms for monitoring drought stress in maize, offering a valuable theoretical foundation and practical reference for non-destructive crop physiological monitoring in precision agriculture.Open AccessEstimation of Chlorophyll and Water Content in Maize Leaves Under Drought Stress Based on VIS/NIR Spectroscopy
Maize (Zea mays) is a key crop, with its growth impacted by drought stress. Accurate, non-destructive assessment of drought severity is crucial for precision agriculture. VIS/NIR reflectance spectroscopy is widely used for estimating plant parameters and detecting stress. However, the relationship between key parameters—such as chlorophyll and water content—and VIS/NIR spectra under drought conditions in maize remains unclear, lacking comprehensive models and validation. This study aims to develop a non-destructive and accurate method for predicting chlorophyll and water content in maize leaves under drought stress using VIS/NIR spectroscopy. Specifically, maize leaf reflectance spectra were collected under varying drought stress conditions, and the effects of different spectral preprocessing methods, dimensionality reduction techniques, and machine learning algorithms were evaluated. An optimal data processing pipeline was systematically established and deployed on an edge computing unit to enable rapid, non-destructive prediction of chlorophyll and water content in maize leaves. The experimental results demonstrated that the combination of stepwise regression (SR) for feature selection and a stacking regression model achieved the best performance for chlorophyll content prediction ( = 0.8740, RMSEp = 0.2768). For leaf water content prediction, random forest (RF) feature selection combined with a stacking model yielded the highest accuracy ( = 0.7626, RMSEp = 4.12%). This study confirms the effectiveness and potential of integrating VIS/NIR spectroscopy with machine learning algorithms for monitoring drought stress in maize, offering a valuable theoretical foundation and practical reference for non-destructive crop physiological monitoring in precision agriculture. - 23 September 2025
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 162
A lightweight citrus detection and counting method based on deep learning model
- 23 September 2025
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 162
A lightweight citrus detection and counting method based on deep learning model
- Open Access
- 5 July 2025
- Published by Elsevier in Plant Phenomics
- Vol. 7 (3)
FSEA: Incorporating domain-specific prior knowledge for few-shot weed detection
- 5 July 2025
- Published by Elsevier in Plant Phenomics
- Vol. 7 (3)
Open AccessFSEA: Incorporating domain-specific prior knowledge for few-shot weed detection
- Open Access
A Multi-Strategy Visual SLAM System for Motion Blur Handling in Indoor Dynamic Environments
Typical SLAM systems adhere to the assumption of environment rigidity, which limits their functionality when deployed in the dynamic indoor environments commonly encountered by household robots. Prevailing methods address this issue by employing semantic information for the identification and processing of dynamic objects in scenes. However, extracting reliable semantic information remains challenging due to the presence of motion blur. In this paper, a novel visual SLAM algorithm is proposed in which various approaches are integrated to obtain more reliable semantic information, consequently reducing the impact of motion blur on visual SLAM systems. Specifically, to accurately distinguish moving objects and static objects, we introduce a missed segmentation compensation mechanism into our SLAM system for predicting and restoring semantic information, and depth and semantic information is then leveraged to generate masks of dynamic objects. Additionally, to refine keypoint filtering, a probability-based algorithm for dynamic feature detection and elimination is incorporated into our SLAM system. Evaluation experiments using the TUM and Bonn RGB-D datasets demonstrated that our SLAM system achieves lower absolute trajectory error (ATE) than existing systems in different dynamic indoor environments, particularly those with large view angle variations. Our system can be applied to enhance the autonomous navigation and scene understanding capabilities of domestic robots.
Open AccessA Multi-Strategy Visual SLAM System for Motion Blur Handling in Indoor Dynamic Environments
Typical SLAM systems adhere to the assumption of environment rigidity, which limits their functionality when deployed in the dynamic indoor environments commonly encountered by household robots. Prevailing methods address this issue by employing semantic information for the identification and processing of dynamic objects in scenes. However, extracting reliable semantic information remains challenging due to the presence of motion blur. In this paper, a novel visual SLAM algorithm is proposed in which various approaches are integrated to obtain more reliable semantic information, consequently reducing the impact of motion blur on visual SLAM systems. Specifically, to accurately distinguish moving objects and static objects, we introduce a missed segmentation compensation mechanism into our SLAM system for predicting and restoring semantic information, and depth and semantic information is then leveraged to generate masks of dynamic objects. Additionally, to refine keypoint filtering, a probability-based algorithm for dynamic feature detection and elimination is incorporated into our SLAM system. Evaluation experiments using the TUM and Bonn RGB-D datasets demonstrated that our SLAM system achieves lower absolute trajectory error (ATE) than existing systems in different dynamic indoor environments, particularly those with large view angle variations. Our system can be applied to enhance the autonomous navigation and scene understanding capabilities of domestic robots.
- 2 December 2024
- Published by Elsevier in Computers and Electronics in Agriculture
- Vol. 229
Robust crop row detection using discrete Fourier transform (DFT) for vision-based in-field navigation
- 2 December 2024
- Published by Elsevier in Computers and Electronics in Agriculture
- Vol. 229
Robust crop row detection using discrete Fourier transform (DFT) for vision-based in-field navigation
- Open Access
- 7 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 162122-162130
A Fruit-Tree Mapping System for Semi-Structured Orchards Based on Multi-Sensor-Fusion SLAM
Robots have been increasingly applied in orchard management, with fruit tree localization and mapping being crucial for precision and automation. Existing methods require real-time kinematic (RTK) global navigation satellite system (GNSS) receivers. However, signal strength is diminished in orchards due to being occluded by tree canopies. Therefore, this study proposes a novel system for global fruit tree positioning and mapping in semi-structured orchards based on multi-sensor fusion simultaneous localization and mapping (SLAM), without dependence of positioning signals from satellites. As the specific contributions, firstly, a SLAM system which fuse a camera, an inertial measurement unit (IMU), and a light detection and ranging (LiDAR) sensorwas constructed to improve the accuracy of robot odometry estimation in complex orchard environments. Secondly, a fruit tree localization algorithm was developed to localize the fruit trees around the robot using both images and LiDAR point clouds, after which the global positions of the detected fruit trees were optimized using the SLAM-derived robot pose real-time. Our method has been extensively evaluated in real pear and persimmon orchards. The results showed that the average root mean square error (RMSE) for odometry was 6.04 cm and 14.59 cm in pear and persimmon orchards, respectively, and the positioning errors of the detected fruit trees were 0.57 cm and 3.23 cm, respectively. The results indicate that our method can achieve accurate positioning and mapping of fruit trees in complex environments and build fruit-tree maps.
- 7 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 162122-162130
Open AccessA Fruit-Tree Mapping System for Semi-Structured Orchards Based on Multi-Sensor-Fusion SLAM
Robots have been increasingly applied in orchard management, with fruit tree localization and mapping being crucial for precision and automation. Existing methods require real-time kinematic (RTK) global navigation satellite system (GNSS) receivers. However, signal strength is diminished in orchards due to being occluded by tree canopies. Therefore, this study proposes a novel system for global fruit tree positioning and mapping in semi-structured orchards based on multi-sensor fusion simultaneous localization and mapping (SLAM), without dependence of positioning signals from satellites. As the specific contributions, firstly, a SLAM system which fuse a camera, an inertial measurement unit (IMU), and a light detection and ranging (LiDAR) sensorwas constructed to improve the accuracy of robot odometry estimation in complex orchard environments. Secondly, a fruit tree localization algorithm was developed to localize the fruit trees around the robot using both images and LiDAR point clouds, after which the global positions of the detected fruit trees were optimized using the SLAM-derived robot pose real-time. Our method has been extensively evaluated in real pear and persimmon orchards. The results showed that the average root mean square error (RMSE) for odometry was 6.04 cm and 14.59 cm in pear and persimmon orchards, respectively, and the positioning errors of the detected fruit trees were 0.57 cm and 3.23 cm, respectively. The results indicate that our method can achieve accurate positioning and mapping of fruit trees in complex environments and build fruit-tree maps.
- Open Access
- 24 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 58703-58714
Agile-RRT*: A Faster and More Robust Path Planner With Enhanced Initial Solution and Convergence Rate in Complex Environments
Path planning is a critical process in mobile robot navigation. Sampling-based path planning algorithms represented by Rapidly Exploring Random Tree star (RRT*) have gained widespread adoption due to their asymptotic optimality and proven efficiency. However, when applied to intricate environments, characterized by narrow passages and cluttered obstacles, these algorithms encounter challenges in both the initial solution generation and the convergence towards the optimal path, mainly caused by the inefficient sampling strategy, thereby impeding its overall effectiveness. To address these limitations, we introduced Agile-RRT* (A-RRT*), an advancement of RRT* algorithm. Our contributions are twofold: firstly, we introduce an adaptive goal-biased sampling strategy, which employs an adaptive principle for determining the step size on the basis of the goal-biased strategy. This avoids getting trapped in local minima and enhances the efficiency of the initial solution generation. Secondly, we introduce a path optimization approach using a secondary tree and subset-informed sampling, to accelerate the convergence toward the optimal path. It optimizes the path by gradually shrinking the designed elliptical planning space around local states, which effectively narrows down the search space. The experimental results demonstrated that the proposed A-RRT* diminishes the initial solution search time by 71.00% and the sub-optimal solution search time by 82.86% in comparison to RRT*. The A-RRT* exhibits superior performance over RRT*, Informed-RRT*, P-RRT* and Quick-RRT* in terms of soundness and efficiency in narrow and intricate environments. This method could expedite efficient motion planning for drones and mobile robots in complex environments.
- 24 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 58703-58714
Open AccessAgile-RRT*: A Faster and More Robust Path Planner With Enhanced Initial Solution and Convergence Rate in Complex Environments
Path planning is a critical process in mobile robot navigation. Sampling-based path planning algorithms represented by Rapidly Exploring Random Tree star (RRT*) have gained widespread adoption due to their asymptotic optimality and proven efficiency. However, when applied to intricate environments, characterized by narrow passages and cluttered obstacles, these algorithms encounter challenges in both the initial solution generation and the convergence towards the optimal path, mainly caused by the inefficient sampling strategy, thereby impeding its overall effectiveness. To address these limitations, we introduced Agile-RRT* (A-RRT*), an advancement of RRT* algorithm. Our contributions are twofold: firstly, we introduce an adaptive goal-biased sampling strategy, which employs an adaptive principle for determining the step size on the basis of the goal-biased strategy. This avoids getting trapped in local minima and enhances the efficiency of the initial solution generation. Secondly, we introduce a path optimization approach using a secondary tree and subset-informed sampling, to accelerate the convergence toward the optimal path. It optimizes the path by gradually shrinking the designed elliptical planning space around local states, which effectively narrows down the search space. The experimental results demonstrated that the proposed A-RRT* diminishes the initial solution search time by 71.00% and the sub-optimal solution search time by 82.86% in comparison to RRT*. The A-RRT* exhibits superior performance over RRT*, Informed-RRT*, P-RRT* and Quick-RRT* in terms of soundness and efficiency in narrow and intricate environments. This method could expedite efficient motion planning for drones and mobile robots in complex environments.
- 4 April 2024
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 210, 282-299
Plant-Denoising-Net (PDN): A plant point cloud denoising network based on density gradient field learning
- 4 April 2024
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 210, 282-299
Plant-Denoising-Net (PDN): A plant point cloud denoising network based on density gradient field learning