About / Resume

Master's student in Electronic Information at Beihang University, focusing on reinforcement learning and multi-agent collaborative control.

Location

Beijing, China

Target Role

AI Agent Algorithm Engineer / Reinforcement Learning Engineer

Master's student in Electronic Information at Beihang University, focusing on reinforcement learning, multi-agent collaboration, and autonomous mission planning. Experienced in both algorithm research and engineering delivery for complex decision-making scenarios.

Education

Beihang University (985)

Sep 2024 - Present

School of Automation Science and Electrical Engineering · M.Eng. in Electronic Information

  • Advisor: Prof. Rui Zhou, National Key Defense Laboratory of Aircraft Control Integration.
  • Research focus: machine learning, reinforcement learning, multi-agent control, and autonomous systems.
  • Leadership: Vice Chair of Graduate Student Union (University), Chair at School level.

Beijing Forestry University (211)

Sep 2020 - Jun 2024

School of Automation · B.Eng. in Automation (AI)

  • Honors: National Scholarship, Beijing Outstanding Graduate, Baosteel Scholarship, First-Class Academic Scholarship.

Research Projects

Large-Scale UAV Swarm Task Planning and Formation Maintenance

Sep 2024 - Dec 2024

Aerospace Innovation Institute · Technical Lead

  • Responsibility: led core algorithm design for target assignment and formation maintenance.
  • Method: developed dynamic assignment, 3D formation control, and formation reconfiguration algorithms.
  • Result: enabled fast rallying, stable formation, and flexible reconfiguration for large swarms.

UAV Swarm Mission Planning

May 2024 - Present

China Aerospace Science & Industry Corp. Second Academy · Technical Lead

  • Responsibility: designed clustering, rally strategy, and anti-collision mechanisms.
  • Method: combined guidance-line tracking with formation restructuring and obstacle avoidance.
  • Result: delivered a real-time oriented collaborative planning solution for UAV swarms.

Data-Driven Intelligent Guidance Methods

Oct 2024 - Dec 2024

China Aerospace Science and Technology Corp. First Academy · Technical Lead

  • Responsibility: built fast prediction models for trajectory and missile reachability.
  • Method: parallel data generation and cleaning pipeline with multiple neural architectures.
  • Result: kept average reachability error within 0.5%.

LLM + Reinforcement Learning for Multi-UAV Collaborative Planning

Nov 2025 - Present

Master's Thesis

  • Responsibility: studied collaborative decision-making and resource scheduling for dynamic/static scenarios.
  • Method: proposed both tightly-coupled optimization and hierarchical decoupled planning models.
  • Result: balanced global optimization quality and real-time constraints under complex conditions.

Internships & Practice

Xiaomi

Jan 2026 - Mar 2026

AI Agent Algorithm Engineer Intern

  • Responsibility: researched and validated game-scene AI Agent capabilities.
  • Method: optimized context modeling, memory mechanisms, and multi-turn decision chains on MIMO base models.
  • Result: improved player behavior simulation and informed device strategy optimization.

ByteDance E-commerce

Aug 2025 - Sep 2025

Multimodal LLM Application Intern

  • Responsibility: explored multimodal LLM applications in Douyin live-streaming scenarios.
  • Method: led large-scale dataset construction for next-generation interactive experiences.
  • Result: validated technical feasibility of multimodal understanding in real-time interaction workflows.

Feishu

Jan 2024 - Mar 2024

Computer Vision Developer (University-Industry Program)

  • Responsibility: participated in CV algorithm iteration for visual understanding and structured extraction.
  • Method: built an automated loop for data cleaning, model training, and evaluation.
  • Result: improved data quality and model iteration efficiency with stronger product alignment.

Graduate Student Union at Beihang

Jul 2025 - Present

Vice Chair / School-level Chair

  • Responsibility: coordinated large-scale events and university-industry collaboration initiatives.
  • Result: partnered with major tech firms and secured approximately CNY 70,000 in sponsorships.

Publications

Context-Aware Relational Learning for Cooperative UAV Formation

First Author

Journal of Beijing Institute of Technology (EI)

  • Proposed CORAL, a multi-agent deep RL framework with contextual awareness and relational learning.
  • Improved cooperative exploration efficiency and teammate-intent inference under sparse rewards.

A Dual-Layer Deep RL Method for Multi-UAV Collaborative Decision and Planning

First Author

Information and Control (EI, Chinese Core Journal)

  • Proposed DAP-DRL, a dual-layer decoupled framework for task assignment and path planning.
  • Designed a three-stage collaborative training strategy for stable coupled optimization.

Competitions

MCM/ICM

Feb 2023

Finalist · Team Leader

China Undergraduate Mathematical Contest in Modeling

Oct 2022

Second Prize (Beijing Region) · Team Member

National Intelligent Vehicle Competition (16th)

Jul 2022

Second Prize (North China) · Team Leader

Forest Fire Smoke Detection System

Oct 2022 - May 2023

First Prize, Beijing Challenge Cup · Team Leader

Skills

C++PythonMATLABPyTorchROSReinforcement LearningMachine LearningMulti-Agent SystemsPath PlanningAI Agent