Hi! I'm Qiyuan LIU.

For me, control systems and robotic design are more than technical endeavors—they are keys to realizing the future. From agriculture to healthcare, from industrial automation to everyday human life, these technologies have the potential to reshape how we interact with the world. Currently, as a Master's student at UC Berkeley, I am immersing myself in advanced studies and hands-on projects in robotics and autonomous systems. This environment not only sharpens my technical skills but also expands my perspective on how engineering can address global challenges.

Looking ahead, I aim to deepen my work at the intersection of control systems and robotic design, constantly pushing the boundaries of technology. By combining theoretical knowledge with real-world needs, I hope to collaborate with like-minded individuals to drive the development of next-generation intelligent systems.

Resume Contact

About me

In my exploration of mechanical engineering and robotics, I have always been committed to finding a balance between technology and real-world needs. Engineering is not just about optimizing existing technologies but also about the art of solving complex problems. From developing reinforcement learning-controlled robotic arms to designing algorithms for multi-robot task allocation, my work goes beyond the application of tools, focusing instead on the logic and potential behind the systems. Every experience has deepened my understanding that innovation stems not only from technical expertise but also from insight into the essence of a problem. In these projects, I strive not only for efficient and precise solutions but also to infuse greater flexibility and adaptability into the technology, ensuring it serves complex and dynamic scenarios effectively.

Location:California Bay Area, US
Interests:photography, traveling, violin
Study:University of California, Berkeley
Employment:Open to Work!

Education

University of California, Berkeley

Masters in Engineering, Mechanical Engineering•August 2024 - Present

Concentrated in Control of Robotics & Autonomous Systems.

Awarded UC Berkeley Eaton-Hachigian Fellowship.

Nanyang Technological University, Singapore

Bachelor of Egineering, Mechanical Engineering•August 2020 - July 2024

Specialized in Robotics and Mechatronics, Graduated with Honors (Highest Distinction).

Experience

Robotic Engineer

Satellite Research Center (SaRC)•December 2022 - May 2023, 6 Months Internship

Implemented motion planning for the UR5 robotic arm using MoveIt with TrajOpt for global trajectory and Cartesian Path for precise end-effector motion. Resulted in improved execution efficiency, and reduced mechanical wear in industrial automation tasks. Deployed a vision-based pushing and grasping system on the UR5 robotic arm, training a self-supervised deep reinforcement learning model (VPG) to enhance object manipulation efficiency, resulting in improved grasp success rates in cluttered environments.

Mechanical Engineer

Surbana Jurong - NTU Corporate Lab•May 2022 - December 2022, 8 Months

Deployed LiDAR SLAM-based indoor navigation, with LIO-SAM for LiDAR-IMU fusion. Integrating A* + DWA for real-time navigation and obstacle avoidance. Accelerated the iScan2BIM workflow by one-third and enabled more efficient indoor navigation. Designed and 3D printed low-friction TPU wheels and flexible chassis components, optimizing tread patterns, wheel geometry, and structural damping to reduce vibration and improve stability in differential drive ground vehicles.

Check out some of my Projects!

Behavior Imitation for Manipulator Control with Deep Reinforcement Learning

Motion imitation is a challenging task in robotics and computer vision, typically requiring expert data obtained through MoCap devices. However, acquiring such data is resource-intensive in terms of financial resources, manpower, and time. Our project addresses this challenge by proposing a novel model that simplifies Motion Imitation into a prediction problem of joint angle values in reinforcement learning.

Intelligence Scan to Building Information Model (iScan2BIM)

Please click this page to visit the official iScan2BIM website. This project was my first exposure to industrial robotics, where I worked as a research assistant. My contributions included assisting in the development of indoor navigation algorithms and path planning, delving deeply into debugging the ROS system, and designing a multi-agent simulation framework.

Weighted Voronoi Cell-based Task Allocation for Collaborative Objects Retrieval

Task allocation is a critical component of multi-robot collaboration. Among various approaches, space partitioning methods based on Voronoi cells (VSP) offer a straightforward and effective way to assign tasks. However, traditional VSP-based task allocation methods often lack adaptability to dynamic changes in tasks and environments. To address this limitation, our model introduces a task allocation algorithm that leverages vehicle-centered Voronoi region partitioning. By dynamically adjusting the weights of Voronoi cells, this method adapts to environmental and task changes in real time. As a result, it achieved a 57.14% increase in task efficiency, and a 21.71% improvement in task distribution balance.

Autonomous Tennis Ball Fetcher

A self-designed mechatronics system capable of fully automated searching and retrieving tennis balls within a designated area.

Get in touch!

Whether you have a project in mind, need collaboration, or are looking for a dedicated Robotics Engineer, feel free to reach out. I’m always open to new opportunities and connections in the robotics and control systems field.

Email: qiyuann@berkeley.edu
Instagram: @qiyuuuan
Github: iQiyuan
Location: University of California, Berkeley, CA