A5072284152- Robotic Path Planning Algorithms
- Robotics and Sensor-Based Localization
- Topic Modeling
- Image Enhancement Techniques
- Sentiment Analysis and Opinion Mining
- Color Science and Applications
- Control and Dynamics of Mobile Robots
- Recommender Systems and Techniques
- Optical Imaging and Spectroscopy Techniques
- Image and Object Detection Techniques
- Human Pose and Action Recognition
- Infrared Target Detection Methodologies
- Advanced Computational Techniques and Applications
- Reinforcement Learning in Robotics
- Video Surveillance and Tracking Methods
- Guidance and Control Systems
- Big Data and Business Intelligence
- Natural Language Processing Techniques
- Human Motion and Animation
- Traffic Prediction and Management Techniques
- Image Processing and 3D Reconstruction
- Advanced Research in Science and Engineering
- Optimization and Search Problems
- Information and Cyber Security
- Advanced Vision and Imaging
Zhejiang University
2016-2025
State Key Laboratory of Modern Optical Instruments
2016-2017
Peking University
2009
The burgeoning interest in developing Large Language Models (LLMs) with up to trillion parameters has been met concerns regarding resource efficiency and practical expense, particularly given the immense cost of experimentation. This scenario underscores importance exploring potential Small (SLMs) as a resource-efficient alternative. In this context, we introduce MiniCPM, specifically 1.2B 2.4B non-embedding parameter variants, not only excel their respective categories but also demonstrate...
Catching high-speed targets in the flight is a complex and typically highly dynamic task. However, existing methods require manual setting of catching height or time, resulting lack adaptability flexibility cannot deal with multiple targets. To bridge this gap, we propose planning-with-decision scheme called catch planner. For sequential decision making, lightweight policy search method based on deep reinforcement learning proposed. It jointly trained motion planning decoupled from physics...
Recent advances in Large Language Models (LLMs) have demonstrated significant potential the field of Recommendation Systems (RSs). Most existing studies focused on converting user behavior logs into textual prompts and leveraging techniques such as prompt tuning to enable LLMs for recommendation tasks. Meanwhile, research interest has recently grown multimodal systems that integrate data from images, text, other sources using modality fusion techniques. This introduces new challenges...
In operations, light reflected from biological tissue can be used for disease detection. this paper, we a visual entropy evaluation method to design the optimal illuminant improve colour discriminability of tissue. The spectral power distribution was obtained by maximising value sample based on human system. experiment, multispectral imaging measure reflectance and clustering conducted extract its features. To verify effectiveness method, simulated images under illuminations with optimised...
Abstract In this paper, we propose a scheme for autonomous exploration in unknown environments using mobile robot. To reduce the storage consumption and speed up search of frontiers, wave-features-based rapidly exploring random tree method, which can inhibit or promote growth sampling trees regionally. Then, prune frontiers with mean shift algorithm use pruned decision-making. avoid repeated exploration, develop decision making method consistency assessment, status robot are explicitly...
In this letter, an aggressive quadrotor ball playing system called Bat is proposed, whose goal to intercept a flying and volley it towards designated target. Aggressive means operates the aggressively balls that are far away hit them distant positions in ways beyond reach of existing methods. The trajectory prediction achieved by integrating forward current position velocity estimates using extended kalman filter, implementing cubic interpolation at time resolution calculate continuous...
Multi-Agent Path Finding (MAPF) has been widely studied in the field of artificial intelligence and multi-agent path search, researchers have made improvements to make such algorithms applicable actual nonholonomic constrained robots. In this paper, collision detection is improved by body-occupied detection, which improves efficiency detection. addition, solve problem low centralized search for multi-robot solutions, reasonable grouping performed reduce number nodes expanded Conflict-based...
Recent advances in Large Language Models (LLMs) have demonstrated significant potential the field of Recommendation Systems (RSs). Most existing studies focused on converting user behavior logs into textual prompts and leveraging techniques such as prompt tuning to enable LLMs for recommendation tasks. Meanwhile, research interest has recently grown multimodal systems that integrate data from images, text, other sources using modality fusion techniques. This introduces new challenges...
For surgical light, it is important to help surgeons easily identify specific tissues. A spectrally controllable lighting environment system was established develop visual and cognitive performance of LED on people in medical environments.
Autonomous target tracking with quadrotors has wide applications in many scenarios, such as cinematographic follow-up shooting or suspect chasing. Target motion prediction is necessary when designing the planner. However, widely used constant velocity rotation assumption can not fully capture dynamics of target. The tracker may fail happens to move aggressively, sudden turn deceleration. In this paper, we propose an intention-aware planner by additionally considering intention enhance safety...
Autonomous unmanned aerial vehicle (UAV) tracking has been widely used in inspection, rescue, follow-up and many other fields. Compared to ground targets, targets are smaller, more flexible possess freedom of motion, which makes the task challenging. In this paper, we propose a vision-based UAV tracker aiming at targets. Firstly, lightweight detection model is deployed for target target's position estimated taking advantage box information camera internal parameters. Then, motion prediction...