Projects

Research and engineering projects that connect 3D perception, collaborative systems, edge AI, and embodied intelligence.

These projects show my research trajectory from resource-constrained intelligent systems toward efficient 3D perception and predictive world modeling. The research projects focus on semantic occupancy, collaborative perception, token communication, and future scene prediction. The engineering projects document earlier hands-on systems work in autonomous driving, robotics, and edge AI.

research

Learning to Merge Tokens for Communication-Efficient Collaborative Occupancy Prediction

Token-based collaborative semantic occupancy prediction with spatio-temporal memory, receiver-driven communication, and content-aware token merging

Bandwidth-Aware Adaptive Token Communication for Collaborative Occupancy Prediction

Information-theoretic framework for communication-efficient multi-agent perception under bandwidth constraints

Collaborative 4D Occupancy World Models with Motion-Aware Token Memory

Collaborative occupancy-based world modeling with motion-aware token memory and future scene prediction

CriticShift: Reliability-Aware Tool-Output Selection

Context-efficient tool-output selection for LLM agents that preserves task-critical evidence under limited context budgets

FedHarness-RB: Reliable Cross-Agent Harness Transfer

Reliable cross-agent transfer of executable harness improvements using target-side validation and selective deployment

QKV-level Token Merging

Attention-level redundancy reduction by exploring token merging inside QKV computation rather than only at representation level

SCMerge: Content-Spatial Consistent Token Merging

Training-free token merging for efficient Transformer inference using semantic similarity and local spatial consistency

Scheduling Optimization Design of IoT Embedded System Based on Improved RMS Algorithm

Real-time task scheduling optimization for IoT embedded systems using an improved RMS algorithm

engineering

Jetson-Based Autonomous Driving System

Embedded autonomous driving system with perception, decision-making, and control on NVIDIA Jetson

Embedded Face Recognition Access Control System

Edge AI access control system with on-device face recognition on STM32

NAO Robot Autonomous Ball Search and Kick System

Embodied AI system for autonomous perception, decision-making, and control on NAO humanoid robots