A5115393946- Traffic control and management
- Autonomous Vehicle Technology and Safety
- Human-Automation Interaction and Safety
- Transportation and Mobility Innovations
- nanoparticles nucleation surface interactions
- Electrostatics and Colloid Interactions
- Nanopore and Nanochannel Transport Studies
- Vehicle Routing Optimization Methods
- Automotive and Human Injury Biomechanics
- VLSI and FPGA Design Techniques
- Vibration and Dynamic Analysis
- Vehicle emissions and performance
- Smart Parking Systems Research
- Marine and Coastal Research
- Optimization and Packing Problems
- Elevator Systems and Control
- Geotechnical Engineering and Underground Structures
- Reinforcement Learning in Robotics
- Maritime Ports and Logistics
- Vehicular Ad Hoc Networks (VANETs)
- Advanced Computational Techniques and Applications
- Muon and positron interactions and applications
- Industrial Technology and Control Systems
- Safety Systems Engineering in Autonomy
- Particle Accelerators and Free-Electron Lasers
Massachusetts Institute of Technology
2021-2024
Xi’an Jiaotong-Liverpool University
2023
Decision Systems (United States)
2023
Lawrence Berkeley National Laboratory
2023
Beijing Research Institute of Uranium Geology
2022
Vehicle routing problems (VRPs) form a class of combinatorial with wide practical applications. While previous heuristic or learning-based works achieve decent solutions on small problem instances up to 100 cities, their performance deteriorates in large problems. This article presents novel learning-augmented local search framework solve large-scale VRP. The method iteratively improves the solution by identifying appropriate subproblems and $\textit{delegating}$ improvement black box...
The construction of marine ranching is a crucial component China’s Blue Granary strategy, yet the fragmented knowledge system in equipment impedes intelligent management and operational efficiency. This study proposes novel graph (KG) framework tailored for equipment, integrating hybrid ontology design, joint entity-relation extraction, graph-based storage: (1) limitations existing KG are obtained through targeted questionnaires diverse users employees; (2)A domain was constructed combined...
Advances in autonomy offer the potential for dramatic positive outcomes a number of domains, yet enabling their safe deployment remains an open problem. This work's motivating question is: In safety-critical settings, can we avoid need to have one human supervise machine at all times? The work formalizes this scalable supervision problem by considering remotely located supervisors and investigating how autonomous agents cooperate achieve safety. article focuses on context vehicles (AVs)...
Autonomous vehicles (AVs) enable more efficient and sustainable transportation systems. Ample studies have shown that controlling a small fraction of AVs can smooth traffic flow mitigate congestion. However, deploying in real-world systems poses challenges due to safety cost concerns. A viable alternative approach be implemented the near future is <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">coarse-grained guidance</i> , where human...
Deep reinforcement learning is a powerful approach to complex decision making. However, one issue that limits its practical application brittleness, sometimes failing train in the presence of small changes environment. This work motivated by empirical observation directly applying an already trained model related task often works remarkably well, also called zero-shot transfer. We take this trick step further consider how systematically select good tasks train, maximizing overall performance...
Autonomous vehicles (AVs) enable more efficient and sustainable transportation systems. Ample studies have shown that controlling a small fraction of AVs can smooth traffic flow mitigate congestion. However, deploying to real-world systems is challenging due transparency safety concerns. An alternative approach deployable in the imminent future human-compatible driving, where human drivers are guided by real-time instructions stabilize traffic. To respect drivers' reaction time $\Delta$,...
Evaluations of Deep Reinforcement Learning (DRL) methods are an integral part scientific progress the field. Beyond designing DRL for general intelligence, task-specific is becoming increasingly prominent real-world applications. In these settings, standard evaluation practice involves using a few instances Markov Decision Processes (MDPs) to represent task. However, many tasks induce large family MDPs owing variations in underlying environment, particularly contexts. For example, traffic...
Autonomous vehicles (AVs) enable more efficient and sustainable transportation systems. Ample studies have shown that controlling a small fraction of AVs can smooth traffic flow mitigate congestion. However, deploying in real-world systems poses challenges due to safety cost concerns. A viable alternative approach be implemented the near future is $\textit{coarse-grained guidance}$, where human drivers are guided by real-time instructions, updated every $\Delta$ seconds, stabilize traffic....
Autonomous vehicles (AVs) hold vast potential to enhance transportation systems by reducing congestion, improving safety, and lowering emissions. AV controls lead emergent traffic phenomena; one such intriguing phenomenon is breaks (rolling roadblocks), where a single efficiently stabilizes multiple lanes through frequent lane switching, similar the highway patrolling officers weaving across during difficult conditions. While previous theoretical studies focus on single-lane mixed-autonomy...
To solve the problem of selling dining and getting most profit in a certain time by two devices, this project is divided into parts, one detection other route planning, these parts are combined to find nearest suitable position plan optimal path from current that point. Firstly, achieve function detecting center points densely populated areas based on work tracing contours point object OpenCV, then select Then out whether road, if yes, will be endpoint, not, use BFS (Breadth-first search)...
The recent development of connected and automated vehicle (CAV) technologies has spurred investigations to optimize dense urban traffic. This paper considers advisory autonomy, in which real-time driving advisories are issued drivers, thus blending the CAV human driver. Due complexity traffic systems, studies coordinating CAVs have resorted leveraging deep reinforcement learning (RL). Advisory autonomy is formalized as zero-order holds, we consider a range hold duration from 0.1 40 seconds....
Abstract Grid-to-rod fretting (GTRF) caused by flow induced vibration is one of the most important mechanisms fuel rod failure in PWRs, which has a strong impact on economy and safety nuclear power plant. To avoid due to GTRF, several verification tests are high costs long term always carried out for new design assemblies. So as deal with disadvantages speed up progress design, progressive analysis method predict GTRF performance developed, systemic verify method. The consists three main...