A5046498809- Robotic Path Planning Algorithms
- Robotics and Sensor-Based Localization
- UAV Applications and Optimization
- Advanced Neural Network Applications
- Target Tracking and Data Fusion in Sensor Networks
- Air Quality Monitoring and Forecasting
- Advanced Chemical Sensor Technologies
- Tactile and Sensory Interactions
- Teleoperation and Haptic Systems
- Indoor and Outdoor Localization Technologies
- Water Quality Monitoring Technologies
- Air Quality and Health Impacts
- Gaze Tracking and Assistive Technology
- Atmospheric chemistry and aerosols
- Evacuation and Crowd Dynamics
- Advanced Memory and Neural Computing
- Reinforcement Learning in Robotics
- Olfactory and Sensory Function Studies
- IoT and Edge/Fog Computing
- Advanced Sensor and Energy Harvesting Materials
- Advanced Manufacturing and Logistics Optimization
- Insect Pheromone Research and Control
- Wireless Signal Modulation Classification
- Autonomous Vehicle Technology and Safety
- Real-Time Systems Scheduling
University Town of Shenzhen
2023-2024
Tsinghua University
2022-2024
University of Electronic Science and Technology of China
2023
Mobile air pollution sensing methods are developed to collect quality data with higher spatial-temporal resolutions. However, existing cannot process the spatially mixed gas samples effectively due highly dynamic temporal and spatial fluctuations experienced by sensor, leading significant measurement deviations. We find an opportunity tackle problem exploring potential patterns from sensor measurements. In light of this, we propose MobiAir, a novel city-scale fine-grained estimation system...
Multi-UAV systems have shown immense potential in handling complex tasks large-scale, dynamic, and cold-start (i.e., limited prior knowledge) scenarios, such as wildfire suppression. Due to the dynamic stochastic environmental conditions, scheduling for sensing fire monitoring) operation suppression) should be executed concurrently enable real-time information collection timely intervention of environment. However, planning inclinations are typically inconsistent evolve over time,...
Hydraulic gantry cranes (hereinafter referred to as “gantry cranes”) are highly susceptible instability during dynamic operations due high-speed unsteady airflow in mountainous and canyon areas, leading safety risks such derailment overturning. Traditional single data-driven or model-driven methods fall short ensuring real-time performance, accuracy, comprehensiveness for the state perception of operations. To overcome this, we draw inspiration from way biomechanics integrates multiple data...
Emergency rescue scenarios are considered to be high-risk scenarios. Using a micro air vehicle (MAV) swarm explore the environment can provide valuable environmental information. However, due absence of localization infrastructure and limited on-board capabilities, it's challenging for low-cost MAV maintain precise localization. In this paper, collaborative system heterogeneous is proposed. This takes full advantage advanced effectively achieve accurate through collaboration. Subsequently,...
Accurate calibration of low-cost sensors is critical for improving their potential in environmental monitoring. State-of-the-art (SOTA) methods based on supervised learning commonly calibrate sensor measurements with ground truth from the immediate past or future. However, these techniques rely heavily labeled data which challenging to obtain real-world scenarios. Thus, this paper introduces CaliFormer, a novel representation model using self-supervised extract time- and spatial-invariant...
Background and motivation. The localization tracking of high-speed dynamic objects find applications across various domains. For example, unmanned aerial vehicles (UAVs) serves a crucial role for regulatory authorities, enabling the swift identification illicitly operated drones.
The landing process of the quadrotors can be affected by disturbance from ground effect when approaching surface. Such a significantly increases chances collision and jittering quadrotors, thereby posing threats to safety both mounted equipment. In light this, we propose SmoothLander, an aerodynamics reinforcement learning-based control system stabilize under influence noise. First, design trajectory for quadrotor in accordance with aerodynamics. Then command generator effectively optimize...
Unforeseen gas leaks can swiftly create a highly flammable atmosphere, which will lead to risks of explosions, health hazards, as well long-term environmental impacts, especially when the is toxic. Hence, localizing source soon possible remains paramount executing subsequent mitigation measures, such shutting off valves and effectively containing or sealing leaks. Assigning human operators engage in these laborious activities poses risk their lives. Deploying autonomous mobile olfactory...
Haptic feedback is essential for dexterous telemanipulation that enables operators to control robotic hands remotely with high skill and precision, mimicking a human hand's natural movement sensation. However, current haptic methods cannot support torque feedback, resulting in object rotation rolling mismatches. The operator must make tedious adjustments these tasks, leading delays, reduced situational awareness, suboptimal task performance. This work presents Bi-directional Momentum-based...
Drone-based rapid and accurate environmental edge detection is highly advantageous for tasks such as disaster relief autonomous navigation. Current methods, using radar or cameras, raise deployment costs burden lightweight drones with high computational demands. In this paper, we propose AirTouch, a system that transforms the ground effect from stability "foe" in traditional flight control views, into "friend" efficient detection. Our key insight analyzing drone sensor readings commands...
Accurately localizing high-speed dynamic objects in 3D space with low latency is crucial for various robotic applications. Current methods face challenges due to extended exposure times and limited sensor resolution, hindering precise object detection localization. Event cameras, known their high temporal resolution asynchronous nature, offer a promising solution. To leverage the potential of event camera, we propose EventTracker, novel framework that integrates depth measurements...
Human-machine interface (HMI) of high degrees freedom (DoF) is one the most critical bases metaverse. The ideal HMI for metaverse should be cheap, robust, customizable, and ergonomically friendly. In light this, we propose a triboelectric nanogenerator (TENG)-based sensing system. We developed low-cost, soft, light, customizable TENG sensor to collect data from human body. then used an artificial neural network (ANN) obtain corresponding motion collected sensory data. effectiveness proposed...
Aiming at the problem that downlink of aerospace measurement and control system has weak anti-dynamic interference performance in dynamic environment, this paper proposes an anti-interference intelligent decision-making algorithm based on Deep Deterministic Policy Gradient (DDPG), applies it to system, which increases system. In paper, structure is given, DDPG evaluation function are designed. The simulation results show compared with rule inference neural network, bit error rate (BER)...