Dawei Wang
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
Research Subjects
Dawei Wang has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Jia Pan
The University of Hong Kong
3 shared publications B. S. Zhang
Department of Electrical & Electronic Engineering, Hong Kong University of Science & Technology, Clear Water Bay, Kowloon, Hong Kong
1 shared publication- Bu Jin
Institute of Automation, Chinese Academy of Sciences (CASIA), Haidian, China
1 shared publication - Chang Chen
The University of Hong Kong
1 shared publication - Chao Yang
Shanghai AI Laboratory, Shanghai, China
1 shared publication - Chengliang Zhong
AIR, Tsinghua University, Haidian, China
1 shared publication
Latest Publications
- Open Access
- 15 October 2026
- Published by Springer Nature in Research Square
Autonomous robotic photography as a creative urban service
Urban service robots are increasingly used for logistics, inspection, and construction, yet their role in cultural and recreational activities remains largely unexplored. We investigate autonomous robotic photography as a creative urban service, addressing a key limitation of existing systems: the difficulty of combining aesthetic judgment with safe, adaptive operation in dynamic public spaces. We present LEAP (Learning Embodied Aesthetics for Photography), a supervised robotic-photography system that integrates learned photography rules with fast state estimation, path planning, and obstacle avoidance. In three supervised urban deployments and 30 controlled photography tasks, LEAP completed all trials without collisions or incomplete-framing failures. It outperformed two robotic-photography baselines, achieving 9.7-19.8\% higher computational aesthetic scores and 8.1-24.5\% higher blinded human aesthetic ratings. LEAP also generalized to previously untrained indoor and outdoor settings and adapted to changing participant poses. These results demonstrate the potential of autonomous robotic photography as an accessible, interactive, and culturally meaningful urban service.
- 15 October 2026
- Published by Springer Nature in Research Square
Open AccessAutonomous robotic photography as a creative urban service
Urban service robots are increasingly used for logistics, inspection, and construction, yet their role in cultural and recreational activities remains largely unexplored. We investigate autonomous robotic photography as a creative urban service, addressing a key limitation of existing systems: the difficulty of combining aesthetic judgment with safe, adaptive operation in dynamic public spaces. We present LEAP (Learning Embodied Aesthetics for Photography), a supervised robotic-photography system that integrates learned photography rules with fast state estimation, path planning, and obstacle avoidance. In three supervised urban deployments and 30 controlled photography tasks, LEAP completed all trials without collisions or incomplete-framing failures. It outperformed two robotic-photography baselines, achieving 9.7-19.8\% higher computational aesthetic scores and 8.1-24.5\% higher blinded human aesthetic ratings. LEAP also generalized to previously untrained indoor and outdoor settings and adapted to changing participant poses. These results demonstrate the potential of autonomous robotic photography as an accessible, interactive, and culturally meaningful urban service.
- Top Cited Papers
- 23 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (9) , 7827-7834
GaussianGrasper: 3D Language Gaussian Splatting for Open-Vocabulary Robotic Grasping
Constructing a 3D scene capable of accommodating open-ended language queries, is a pivotal pursuit in the domain of robotics, which facilitates robots in executing object manipulations based on human language directives. To achieve this, some research efforts have been dedicated to the development of language-embedded implicit fields. However, implicit fields (e.g. NeRF) encounter limitations due to the necessity of taking images from a larger number of viewpoints for reconstruction, coupled with their inherent inefficiencies in inference. Furthermore, these methods directly distill patch-level 2D features, leading to ambiguous segmentation boundaries. Thus, we present the GaussianGrasper , which uses 3D Gaussian Splatting (3DGS) to explicitly represent the scene as a set of Gaussian primitives and is capable of real-time rendering. Our approach takes RGB-D images from limited viewpoints as input and uses an Efficient Feature Distillation (EFD) module that employs contrastive learning to efficiently distill 2D language embeddings and constraint consistency of feature embeddings. With the reconstructed geometry of the Gaussian field, our method enables the pre-trained grasping model to generate collision-free grasp pose candidates. Furthermore, we propose a normal-guided grasp module to select the best grasp pose. Through comprehensive real-world experiments, we demonstrate that GaussianGrasper enables robots to accurately locate and grasp objects according to language instructions, providing a new solution for language-guided grasping tasks.
- 23 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (9) , 7827-7834
Top Cited PapersGaussianGrasper: 3D Language Gaussian Splatting for Open-Vocabulary Robotic Grasping
Constructing a 3D scene capable of accommodating open-ended language queries, is a pivotal pursuit in the domain of robotics, which facilitates robots in executing object manipulations based on human language directives. To achieve this, some research efforts have been dedicated to the development of language-embedded implicit fields. However, implicit fields (e.g. NeRF) encounter limitations due to the necessity of taking images from a larger number of viewpoints for reconstruction, coupled with their inherent inefficiencies in inference. Furthermore, these methods directly distill patch-level 2D features, leading to ambiguous segmentation boundaries. Thus, we present the GaussianGrasper , which uses 3D Gaussian Splatting (3DGS) to explicitly represent the scene as a set of Gaussian primitives and is capable of real-time rendering. Our approach takes RGB-D images from limited viewpoints as input and uses an Efficient Feature Distillation (EFD) module that employs contrastive learning to efficiently distill 2D language embeddings and constraint consistency of feature embeddings. With the reconstructed geometry of the Gaussian field, our method enables the pre-trained grasping model to generate collision-free grasp pose candidates. Furthermore, we propose a normal-guided grasp module to select the best grasp pose. Through comprehensive real-world experiments, we demonstrate that GaussianGrasper enables robots to accurately locate and grasp objects according to language instructions, providing a new solution for language-guided grasping tasks.
- 21 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (2) , 1981-1989
Large-Scale Mixed Traffic Control Using Dynamic Vehicle Routing and Privacy-Preserving Crowdsourcing
Controlling and coordinating urban traffic flow through robot vehicles (RVs) is emerging as a novel transportation paradigm for the future. While this approach garners growing attention from researchers and practitioners, effectively managing and coordinating large-scale mixed traffic remains a challenge. We introduce an effective framework for large-scale mixed traffic control via privacy-preserving crowdsourcing and dynamic vehicle routing. Our framework consists of three modules: 1) a privacy-protecting crowdsensing method; 2) a graph propagation-based traffic forecasting method; and 3) a privacy-preserving route selection mechanism. We evaluate our framework using a real-world road network. The results show that our framework accurately forecasts traffic flow, efficiently mitigates network-wide RV shortage issue, and coordinates large-scale mixed traffic. Compared to other baseline methods, our framework not only reduces the RV shortage issue up to 69.4% but also reduces the average waiting time of all vehicles in the network up to 27%.
- 21 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 11 (2) , 1981-1989
Large-Scale Mixed Traffic Control Using Dynamic Vehicle Routing and Privacy-Preserving Crowdsourcing
Controlling and coordinating urban traffic flow through robot vehicles (RVs) is emerging as a novel transportation paradigm for the future. While this approach garners growing attention from researchers and practitioners, effectively managing and coordinating large-scale mixed traffic remains a challenge. We introduce an effective framework for large-scale mixed traffic control via privacy-preserving crowdsourcing and dynamic vehicle routing. Our framework consists of three modules: 1) a privacy-protecting crowdsensing method; 2) a graph propagation-based traffic forecasting method; and 3) a privacy-preserving route selection mechanism. We evaluate our framework using a real-world road network. The results show that our framework accurately forecasts traffic flow, efficiently mitigates network-wide RV shortage issue, and coordinates large-scale mixed traffic. Compared to other baseline methods, our framework not only reduces the RV shortage issue up to 69.4% but also reduces the average waiting time of all vehicles in the network up to 27%.
- 14 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 24 (8) , 8222-8235
Fuel Rate Prediction for Heavy-Duty Trucks
Fuel cost contributes significantly to the high operation cost of heavy-duty trucks. Developing fuel rate prediction models is the cornerstone of fuel consumption optimization approaches for heavy-duty trucks. However, limited by accurate features directly related to the truck’s fuel consumption, state-of-the-art models show poor performance and are rarely deployed in practice. In this paper, we use the truck’s engine management system (EMS) and Instant Fuel Meter (IFM) to collect a three-month dataset during the period of December 2019 to June 2020. Seven prediction models, including linear regression, polynomial regression, MLP, CNN, LSTM, CNN-LSTM, and AutoML, are investigated and evaluated to predict real-time fuel rate. The evaluation results show that the EMS and IFM dataset help to improve the coefficient of determination of traditional linear/polynomial models from 0.87 to 0.96, while learning-based approach AutoML improves the coefficient of determination to attain 0.99. Besides, we explore the actual deployment of fuel rate prediction with transfer learning and path planning for autonomous driving.
- 14 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 24 (8) , 8222-8235
Fuel Rate Prediction for Heavy-Duty Trucks
Fuel cost contributes significantly to the high operation cost of heavy-duty trucks. Developing fuel rate prediction models is the cornerstone of fuel consumption optimization approaches for heavy-duty trucks. However, limited by accurate features directly related to the truck’s fuel consumption, state-of-the-art models show poor performance and are rarely deployed in practice. In this paper, we use the truck’s engine management system (EMS) and Instant Fuel Meter (IFM) to collect a three-month dataset during the period of December 2019 to June 2020. Seven prediction models, including linear regression, polynomial regression, MLP, CNN, LSTM, CNN-LSTM, and AutoML, are investigated and evaluated to predict real-time fuel rate. The evaluation results show that the EMS and IFM dataset help to improve the coefficient of determination of traditional linear/polynomial models from 0.87 to 0.96, while learning-based approach AutoML improves the coefficient of determination to attain 0.99. Besides, we explore the actual deployment of fuel rate prediction with transfer learning and path planning for autonomous driving.
- 23 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 6 (3) , 4520-4527
VR-ORCA: Variable Responsibility Optimal Reciprocal Collision Avoidance
As one of the most popular multi-agent path planning approaches, the optimal reciprocal collision avoidance (ORCA) algorithm assumes that each agent takes half the responsibility for collision avoidance. However, due to the asymmetric situation faced by adjacent agents, they are expected to take different responsibilities for collision avoidance to improve the entire crowd's navigation performance. Thus, in this letter, we propose the variable responsibility optimal reciprocal collision avoidance (VR-ORCA) algorithm, which relaxes the original assumption in ORCA and only requires the responsibilities of a pair of agents sum to one. In particular, the responsibility division between a pair of nearby agents is determined independently for each agent by minimizing a cost function involving their common neighbors. We validate our approach on a variety of simulated benchmarks and the results demonstrate that our novel method is beneficial in reducing the collision probability, travel time and distance of ORCA.
- 23 March 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 6 (3) , 4520-4527
VR-ORCA: Variable Responsibility Optimal Reciprocal Collision Avoidance
As one of the most popular multi-agent path planning approaches, the optimal reciprocal collision avoidance (ORCA) algorithm assumes that each agent takes half the responsibility for collision avoidance. However, due to the asymmetric situation faced by adjacent agents, they are expected to take different responsibilities for collision avoidance to improve the entire crowd's navigation performance. Thus, in this letter, we propose the variable responsibility optimal reciprocal collision avoidance (VR-ORCA) algorithm, which relaxes the original assumption in ORCA and only requires the responsibilities of a pair of agents sum to one. In particular, the responsibility division between a pair of nearby agents is determined independently for each agent by minimizing a cost function involving their common neighbors. We validate our approach on a variety of simulated benchmarks and the results demonstrate that our novel method is beneficial in reducing the collision probability, travel time and distance of ORCA.
- 18 May 2006
- Published by Wiley in Physica Status Solidi (a)
- Vol. 203 (7) , 1712-1715
Fabrication of suspended GaN microstructures using GaN-on-patterned-silicon (GPS) technique
- 18 May 2006
- Published by Wiley in Physica Status Solidi (a)
- Vol. 203 (7) , 1712-1715
Fabrication of suspended GaN microstructures using GaN-on-patterned-silicon (GPS) technique