@article{zhang_he_wu_quesada_xiang_2024, title={Development of a bionic hexapod robot with adaptive gait and clearance for enhanced agricultural field scouting}, volume={11}, ISSN={["2296-9144"]}, DOI={10.3389/frobt.2024.1426269}, abstractNote={High agility, maneuverability, and payload capacity, combined with small footprints, make legged robots well-suited for precision agriculture applications. In this study, we introduce a novel bionic hexapod robot designed for agricultural applications to address the limitations of traditional wheeled and aerial robots. The robot features a terrain-adaptive gait and adjustable clearance to ensure stability and robustness over various terrains and obstacles. Equipped with a high-precision Inertial Measurement Unit (IMU), the robot is able to monitor its attitude in real time to maintain balance. To enhance obstacle detection and self-navigation capabilities, we have designed an advanced version of the robot equipped with an optional advanced sensing system. This advanced version includes LiDAR, stereo cameras, and distance sensors to enable obstacle detection and self-navigation capabilities. We have tested the standard version of the robot under different ground conditions, including hard concrete floors, rugged grass, slopes, and uneven field with obstacles. The robot maintains good stability with pitch angle fluctuations ranging from −11.5° to 8.6° in all conditions and can walk on slopes with gradients up to 17°. These trials demonstrated the robot’s adaptability to complex field environments and validated its ability to maintain stability and efficiency. In addition, the terrain-adaptive algorithm is more energy efficient than traditional obstacle avoidance algorithms, reducing energy consumption by 14.4% for each obstacle crossed. Combined with its flexible and lightweight design, our robot shows significant potential in improving agricultural practices by increasing efficiency, lowering labor costs, and enhancing sustainability. In our future work, we will further develop the robot’s energy efficiency, durability in various environmental conditions, and compatibility with different crops and farming methods.}, journal={FRONTIERS IN ROBOTICS AND AI}, author={Zhang, Zhenghua and He, Weilong and Wu, Fan and Quesada, Lina and Xiang, Lirong}, year={2024}, month={Sep} }
 @article{he_li_zhang_chen_zhang_panthee_meadows_xiang_2024, title={High-Throughput Robotic Phenotyping for Quantifying Tomato Disease Severity Enabled by Synthetic Data and Domain-Adaptive Semantic Segmentation}, volume={12}, ISSN={["1556-4967"]}, DOI={10.1002/rob.22490}, abstractNote={ABSTRACT Plant diseases cause an annual global crop loss of 20%–40%, leading to estimated economic losses of 30–50 billion dollars. Tomatoes are susceptible to more than 200 diseases. Breeding disease‐resistant cultivars is more cost‐effective and environmentally sustainable than the frequent use of pesticides. Traditional breeding methods for disease resistance, relying on direct visual observation to measure disease‐related traits, are time‐consuming, inaccurate, expensive, and require specific knowledge of tomato diseases. High‐throughput disease phenotyping is essential to reduce labor costs, improve measurement accuracy, and expedite the release of new varieties, thereby more effectively identifying disease‐resistant crops. Precision agriculture efforts have primarily focused on detecting diseases on individual tomato leaves under controlled laboratory conditions, neglecting the assessment of disease severity of the entire plant in the field. To address this, we created a synthetic data set using existing field and individual leaf data sets, leveraging a game engine to minimize additional data labeling. Consequently, we developed a customized unsupervised domain‐adaptive tomato disease segmentation algorithm that monitors the entire tomato plant and determines disease severity based on the proportion of affected leaf areas. The system‐derived disease percentages show a high correlation with manually labeled data, evidenced by a correlation coefficient of 0.91. Our research demonstrates the feasibility of using ground robots equipped with deep‐learning algorithms to monitor tomato disease severity under field conditions, potentially accelerating the automation and standardization of whole‐plant disease severity monitoring in tomatoes. This high‐throughput disease phenotyping system can also be adapted to analyze diseases in other crops with similar foliar diseases, such as maize, soybeans, and cotton.}, journal={JOURNAL OF FIELD ROBOTICS}, author={He, Weilong and Li, Xingjian and Zhang, Zhenghua and Chen, Yuxi and Zhang, Jianbo and Panthee, Dilip R. and Meadows, Inga and Xiang, Lirong}, year={2024}, month={Dec} }
 @article{he_gage_rellan-alvarez_xiang_2024, title={Swin-Roleaf: A new method for characterizing leaf azimuth angle in large-scale maize plants}, volume={224}, ISSN={["1872-7107"]}, url={https://doi.org/10.1016/j.compag.2024.109120}, DOI={10.1016/j.compag.2024.109120}, journal={COMPUTERS AND ELECTRONICS IN AGRICULTURE}, author={He, Weilong and Gage, Joseph L. and Rellan-Alvarez, Ruben and Xiang, Lirong}, year={2024}, month={Sep} }