CV

Basic Information

• Name: Qisheng Zhao (赵祺晟)
• Phone: 15279400369 (WeChat ID)
• Email: 2301213177@stu.pku.edu.cn
• Date of Birth: August 2001
• University: Peking University
• Degree: Master’s Student (Currently Enrolled)

Education Background

Master’s Degree

• Period: 2023.09-2026.07
• University: Peking University
• College/Department: School of Advanced Manufacturing and Robotics
• Major/Program: Advanced Manufacturing and Intelligent Systems
• GPA: 3.66/4.00 (Top 5%)
• Main Courses: Artificial Intelligence and Autonomous Robots, Principles of Reinforcement Learning, AI Methods and Applications, Hybrid Control Systems, Advanced Robot Control, Multi-Agent Systems, Python Big Data Analysis Principles and Applications, Data Visualization

Bachelor’s Degree

• Period: 2019.09-2023.07
• University: Tongji University
• College/Department: CDHAW
• Major/Program: Mechatronics Engineering
• GPA: 4.62/5.00 (Top 10%)
• Honors: Excellent Undergraduate Graduate of Tongji University
• Main Courses: Principles of Automatic Control, Signals and Systems, Network and Industrial Communication, System Modeling and Simulation, Python, C++, Linear Algebra
• Scholarships: Phoenix Contact Scholarship (Excellence Award), First-class Scholarship for Excellent Students, Second-class Scholarship for Excellent Students, DAAD Scholarship

Research Project Experience

Multi-robot Reactive Collaborative Planning with Uncertainty Perception for Dynamic Temporal Tasks

• Period: 2024.06-2025.08
• Role: Project Leader
• Description: Focuses on task allocation algorithms for heterogeneous robot systems with complex temporal constraints and mobile target associations, aiming to minimize average task completion time.
• Implementation Plan: Target trajectory prediction using LSTM and Conformal Prediction (CP), task modeling with Linear Temporal Logic (LTL), task allocation using Monte Carlo Tree Search (MCTS) with CP-based metrics and Conditional Value-at-Risk (CVaR), and online adaptation with a rolling horizon strategy.
• My Contribution: Algorithm design and code implementation, comparison of simulation and physical experiment validation, thesis and patent writing.

Optimization of Battle Strategies in Honor of Kings based on VDN Reinforcement Learning Algorithm

• Period: 2025.06-2025.07
• Role: Project Leader
• Description: Develops a reinforcement learning adversarial model for a game scenario with five heterogeneous intelligent agents (heroes) fighting a tyrant, aiming to minimize the tyrant’s remaining health.
• Implementation Plan: Environment setup using Docker and algorithm improvement by redesigning the reward function based on Value Decomposition Networks (VDN), considering tyrant’s health, skill releases, time penalties, and end states.
• My Contribution: Experimental environment configuration, reward function design, model training iterations.

Mujoco Robotic Arm Grasping Simulation Experiment based on GraspNet

• Period: 2025.05-2025.06
• Role: Project Leader
• Description: Replicates and deploys the GraspNet model in Mujoco simulation for robotic arm grasping experiments in multi-object scenarios, verifying GraspNet’s accuracy and robustness.
• Implementation Plan: Mujoco simulation + GraspNet algorithm
• My Contribution: Literature review, Mujoco environment setup, GraspNet model deployment and testing.

Robot Autonomous Exploration in Unknown Maps (Physical Experiment)

• Period: 2024.11-2024.12
• Role: Project Leader
• Description: Completed autonomous exploration, mapping, and navigation using a LIMO small car in a randomly constructed obstacle environment.
• Implementation Plan: Autonomous Exploration: Frontier-based exploration strategy; SLAM: Gmapping. Autonomous Navigation: Voronoi diagram and A*-based path planning + PID-based path tracking.
• My Contribution: Design and development of autonomous navigation module algorithms, integration and debugging of various modules.

Development of Rapid Task Allocation and On-the-Fly Planning Algorithms

• Period: 2024.04-2024.11
• Role: Project Leader
• Description: Designed fast and efficient task allocation algorithms and on-the-fly planning strategies for cluster tasks with complex timing and strict time window constraints, as well as heterogeneous multi-agent systems, capable of handling emergencies (e.g., new task releases, agent damage).
• Implementation Plan: Task Modeling and Analysis: Described tasks using Linear Temporal Logic (LTL), extracted essential spatiotemporal constraints between tasks by computing partial orders. Task Allocation: Employed a branch-and-bound search framework to efficiently explore the feasible solution space, recorded the optimal value with an upper bound, and filtered out unpromising branches with a lower bound.
• My Contribution: Design and code implementation of algorithm solutions, algorithm development and integration, writing of technical and test documentation.

Internship Experience

CogLeap.AI

• Period: 2025.04-2025.09
• Role: Autonomous Exploration and Navigation Module Leader
• Job Description: Responsible for the design and development of the company’s active 3D reconstruction robot’s autonomous exploration and navigation module, achieving efficient exploration, reconstruction, and motion planning.
• Implementation Plan: Autonomous Exploration: 1. Frontier detection in parallel based on Rapidly-exploring Random Tree (RRT); 2. Hierarchical viewpoint sampling and path planning (TARE). Autonomous Navigation: ROS2 Navigation2 package; SLAM: Cartographer.
• My Contribution: Design and development of exploration and navigation algorithm modules, ROS integration and deployment on the robot chassis, experimental testing in complex and diverse scenarios.

ZF China Investment Co., Ltd.

• Period: 2023.02-2023.05
• Role: Software Testing Department Intern
• Job Description: Participated in functional testing and regression verification of in-vehicle software systems, assisted in completing multi-version system functional verification and automated script writing.
• My Contribution: Writing Python-based functional modules and test scripts, writing test manuals and reproduction process documentation.

Achievements

Publications

• Paper: “UMBRELLA: Uncertainty-aware Multi-roBot REactive Coordination under Dynamic Temporal Logic TAsks”. IEEE International Conference on Robotics and Automation (ICRA) (Top robotics conference, first author).

Patents

• “Adaptive Planning for Multi-Robot Uncertainty Perception in Dynamic Tasks”
• “Dynamic Allocation Decision Method and System for Unmanned Swarm Based on Formal Methods”
• “Shortest Time Decision Method and System for Cluster Collaborative Tasks Based on Formal Methods”
• “Digital Collaborative Design and Verification Method, Device, and Storage Medium for Automated Equipment”

Competitions

• Third Prize in Tongji University Physics Competition
• Second Prize in Jiangxi Province High School Mathematics Competition
• Second Prize in Jiangxi Province High School Chemistry Competition

Skills & Expertise

Language Skills

• English CET-6 (559) / IELTS (6.5)
• German TestDaf (19/20)

Technical Skills

• Task Decision Algorithms: Automata, Behavior Trees, Formal Methods
• Autonomous Exploration Algorithms: TARE, ERRT, FUEL
• Motion Planning Algorithms: Dijkstra, A*, RRT, Optimal Transport
• Programming Software: Python, C++, ROS, Navigation2, Linux, LTL2BA

Personal Summary

• Clear Logic, Strong Learning Ability: Multiple research and project experiences have shaped my clear logical thinking and honed my ability to think independently and delve into learning.
• Diligent, Strong Sense of Responsibility: I have participated in multiple group works, and as a class representative, I served as a bridge for communication between teachers and students, possessing rich coordination experience.
• Eager to Try, Strong Curiosity: I actively expand my hobbies, including piano, badminton, table tennis, squash, and street dance.