Tracking the Ground: Online Lidar Identification of Robot-Induced Soil Deformation in Agricultural Environments
本文提出了一种基于激光雷达观测的在线方法，用于量化和估计农业环境中由机器人引起的土壤变形，以实现对土壤状态的实时监测与保护。
0 citationsRead paper
Selected work
Representative Papers
When Obstacles Bend: Modeling Vegetation Deformation in the context of Field Robotics
研究通过结合变形测量和接触力数据，独立于特定机器人来表征植被的内在机械属性，以解决自然环境中自主机器人与植被交互的问题。
0 citationsRead paper
Hydrogen production from blended waste biomass: pyrolysis, thermodynamic-kinetic analysis and AI-based modelling
This study investigates the sustainable hydrogen production potential of two underutilized food-derived biomass wastes—spent coffee grounds and jujube pits—via pyrolysis. Thermogravimetric analysis (TGA/DTG), pyrolysis-micro gas chromatography (Py-Micro GC), and kinetic modeling (KAS, FWO, and Friedman methods) were employed to systematically characterize pyrolytic behavior and H₂ yield for individual feedstocks and three blended formulations (Blend 1–3). An AI-enhanced modeling framework was innovatively developed, integrating long short-term memory (LSTM) neural networks to achieve highly accurate prediction of thermogravimetric profiles (R² = 0.9998) and to elucidate inter-material synergistic effects. Results show that Blend 3 delivers the highest hydrogen yield, while Blend 1 exhibits the lowest apparent activation energy (161.75 kJ/mol). This work establishes a scalable, integrated kinetics–AI optimization paradigm for targeted hydrogen production from waste biomass.
0 citationsRead paper
StixelNExT++: Lightweight Monocular Scene Segmentation and Representation for Collective Perception
To address the high computational cost and low representation efficiency of monocular 3D scene understanding in autonomous driving collective perception, this paper proposes a lightweight monocular 3D scene representation method. Our approach integrates fine-grained 3D Stixel units with a learnable clustering mechanism, enabling semantic-aware adaptive clustering that compresses scene representations while improving object segmentation accuracy. We design a lightweight neural network that takes a single RGB image as input and jointly leverages depth estimation and LiDAR-based self-supervised ground truth to efficiently generate Stixel representations—natively supporting multimodal outputs including point clouds and bird’s-eye-view (BEV) maps. Evaluated on the Waymo Open Dataset within a 30-meter range, our method achieves state-of-the-art performance with only 10 ms inference time per frame, striking an optimal balance among real-time efficiency, accuracy, and compatibility with collaborative perception systems.
0 citationsRead paper
Recent publications
Latest Papers
Tracking the Ground: Online Lidar Identification of Robot-Induced Soil Deformation in Agricultural Environments
本文提出了一种基于激光雷达观测的在线方法，用于量化和估计农业环境中由机器人引起的土壤变形，以实现对土壤状态的实时监测与保护。
0 citationsRead paper
When Obstacles Bend: Modeling Vegetation Deformation in the context of Field Robotics
研究通过结合变形测量和接触力数据，独立于特定机器人来表征植被的内在机械属性，以解决自然环境中自主机器人与植被交互的问题。
0 citationsRead paper
Hydrogen production from blended waste biomass: pyrolysis, thermodynamic-kinetic analysis and AI-based modelling
This study investigates the sustainable hydrogen production potential of two underutilized food-derived biomass wastes—spent coffee grounds and jujube pits—via pyrolysis. Thermogravimetric analysis (TGA/DTG), pyrolysis-micro gas chromatography (Py-Micro GC), and kinetic modeling (KAS, FWO, and Friedman methods) were employed to systematically characterize pyrolytic behavior and H₂ yield for individual feedstocks and three blended formulations (Blend 1–3). An AI-enhanced modeling framework was innovatively developed, integrating long short-term memory (LSTM) neural networks to achieve highly accurate prediction of thermogravimetric profiles (R² = 0.9998) and to elucidate inter-material synergistic effects. Results show that Blend 3 delivers the highest hydrogen yield, while Blend 1 exhibits the lowest apparent activation energy (161.75 kJ/mol). This work establishes a scalable, integrated kinetics–AI optimization paradigm for targeted hydrogen production from waste biomass.
0 citationsRead paper
StixelNExT++: Lightweight Monocular Scene Segmentation and Representation for Collective Perception
To address the high computational cost and low representation efficiency of monocular 3D scene understanding in autonomous driving collective perception, this paper proposes a lightweight monocular 3D scene representation method. Our approach integrates fine-grained 3D Stixel units with a learnable clustering mechanism, enabling semantic-aware adaptive clustering that compresses scene representations while improving object segmentation accuracy. We design a lightweight neural network that takes a single RGB image as input and jointly leverages depth estimation and LiDAR-based self-supervised ground truth to efficiently generate Stixel representations—natively supporting multimodal outputs including point clouds and bird’s-eye-view (BEV) maps. Evaluated on the Waymo Open Dataset within a 30-meter range, our method achieves state-of-the-art performance with only 10 ms inference time per frame, striking an optimal balance among real-time efficiency, accuracy, and compatibility with collaborative perception systems.
0 citationsRead paper