Research Interest
Driving into or parking in indoor environments such as underground parking garages, often facechallenges in navigation since it is difficult to access to the signals from the Global Navigation Satellite Systems (GNSS).
As an essential part of the Advanced Driver Assistance Systems (ADAS) and Autonomous Vehicles (AVs), the environment-aware technology needs to effectively solve the problem of 3Dinformation acquisition of indoor and outdoor environments. According to GNSS and Inertial Measurement Unit (IMU), the high-precision motion trajectory can be constructed to restore 3D LiDAR point clouds. However, once removed from GNSS, such as GNSS-denied indoor/underground environments, such scanning systems are ineffective. As a matter of fact, last mile automatic parking in underground garages need algorithmssuch as simultaneous localization and mapping (SLAM) to get accurate position and semantic information. The algorithm should not only relies on the sensors installed in the vehicle for environmental data analysis but also requires accurate high-definition (HD) maps for precise navigation. These requirements make the theory and algorithm of high-precision 3D reconstruction, semantic segmentation, and target extraction in indoor scenes one of the most attractive research topics.
My current research mainly focuses on the last-mile solution for autonomous driving in GNSS-denied environments, including the target-free automatic self-calibration, feature narration that combines 2D and 3D features to improve relocation accuracy and speed, high-level semantic information extraction and update from point cloud data. In addition, with the multi-sensory mobile system, I am working on SLAM-based 3D mapping and environment perception by fusion of LiDAR point clouds and camera images to get more concise indoor data acquisition and 3D scene restoration. This solution will work toward the improvement of the accuracy, efficiency, and robustness of indoor augmented reality.
Education
- MSc, Electrical and Computer Engineering, Baylor University, USA ,2018
- BSEng, Electrical and Computer Engineering, Baylor University, USA, 2016
Publications
- Zhang D, Gong Z, Chen Y, Zelek J, *Li J, 2019. SLAM-based multi sensor backpack system LiDAR system on GNSS-denied environment,IGARSS 2019, July 28 - Aug 2, Yokohama, Japan(IEEE: EI)
- Gong Z, Lin H, Zhang D. Luo Z, Zelek J, Chen Y, Wang C, *Li J, 2019. A frustum-based probabilistic framework for 3D object detection by fusion of lidar and camera data, ISPRS Journal of Photogrammetry and Remote Sensing, under review.
- Ma L, Chen Z, Li Y, Zhang D, *Li J, Chapman M, 2019 Multispectral airborne laser scanning point-clouds for land cover classification using convolutional neural networks, ISPRS Achieve, 42(2/W13), 79-86, ISPRS Geospatial Week 2019, doi:10.5194/isprs-archives-XLII-2-W13-79-2019.