Sign In

David Liu

Director of Engineering at AeroVironment, Ex-Meta/Facebook

Li Zhang is a knowledgeable software engineer with a Master of Science (M.S.) in Computer Science from the prestigious University of California, Los Angeles, as well as a Bachelor of Science (BS) in Mathematics from The Ohio State University. With a diverse educational background, Li has also studied at Shanghai University of Finance & Economics.

Currently, Li is employed at Facebook as a Software Engineer, where they apply their skills and experience to develop top-notch software solutions for this leading social media platform. Prior to this, Li worked at AppFolio Inc. as a Software Engineer, gaining valuable experience and making meaningful contributions to the company's product line.

At AppFolio Inc., Li also worked as a Software Engineering Intern, further deepening their understanding of software development best practices and gain critical experience in the field. Before this, they were a Graduate Student at UCLA, where they continued to hone their skills and gain expertise in the field of computer science. Before that, they studied at The Ohio State University, earning a degree in Mathematics.

Overall, Li's background highlights their exceptional educational credential and extensive professional experience- making them an asset to any organization regardless of industry.

Highlights

Mar 31 · linkedin

𝐂𝐨𝐦𝐩𝐥𝐞𝐦𝐞𝐧𝐭𝐚𝐫𝐲 𝐍𝐚𝐭𝐮𝐫𝐞 𝐨𝐟 𝐂𝐥𝐚𝐬𝐬𝐢𝐜 𝐀𝐥𝐠𝐨𝐫𝐢𝐭𝐡𝐦𝐬 𝐚𝐧𝐝 𝐋𝐚𝐫𝐠𝐞 𝐌𝐋 𝐌𝐨𝐝𝐞𝐥𝐬 𝐢𝐧 𝐑𝐨𝐛𝐨𝐭𝐢𝐜𝐬 𝐃𝐞𝐯𝐞𝐥𝐨𝐩𝐦𝐞𝐧𝐭

Throughout my career, I've had the privilege of delving into the development of algorithms powering various autonomous systems, including drones, self-driving cars, and RF/optical pointing. These systems all rely on robust algorithms aimed at efficiently navigating from one point to another by taking in sensor inputs and generating control outputs.

𝐂𝐥𝐚𝐬𝐬𝐢𝐜 𝐀𝐩𝐩𝐫𝐨𝐚𝐜𝐡: For over half a century, the development of autonomous systems for spacecraft, aircraft, and beyond has thrived on a foundation of shared algorithmic architecture.  These systems use both traditional filtering and control methods and advanced deep learning techniques for modeling, mapping, perception, state estimation, guidance/planning, and control. The table below provides an overview of the key aspects of each system and highlights both their similarities and differences.

𝐂𝐡𝐚𝐥𝐥𝐞𝐧𝐠𝐞𝐬: It is fascinating to note the classic algorithmic approach based on physics and mathematics has successfully tackled remarkably complex engineering problems. For example, determining one’s precise location using IMU and GNSS signals or controlling unstable systems such as the double inverted pendulum. However, for tasks that are intuitively simple for humans, such as driving a car or doing daily chores, engineers have found the design complexity and handling of all corner cases using classic algorithms to be unmanageable.

𝐓𝐫𝐞𝐧𝐝 𝐨𝐟 𝐮𝐬𝐢𝐧𝐠 𝐋𝐚𝐫𝐠𝐞 𝐌𝐋 𝐌𝐨𝐝𝐞𝐥𝐬: Similar to the Large Language Model and its adaptation to specific applications, the future of robotics algorithm development is poised for a similar paradigm shift. The algorithm architecture for robotics is likely to undergo restructuring, condensing into 3 layers:

1) Classic algorithmic approach for sensor processing and control 2) A unified multi-modal foundational model to handle real-world complexity. It takes a variety of sensor and map data as inputs and generates control targets as outputs. This large robotics model will embed all aspects of intelligence into one cohesive entity, replacing much of the complexity in perception, planning, guidance, and safety. 3) Fine-tuning and augmentation of the unified multi-modal foundational model to suit specific applications and regulatory constraints.

Oct 13 · linkedin

AeroVironment is hiring a Sr Mechanical Engineering Manager in Petaluma, CA to lead a team to work on the Jump 20, a 200 lb 14 hr endurance VTOL fixed-wing UAV, https://lnkd.in/gSZjNUiq

If you are interested, DM me. https://lnkd.in/gK7_QyBm

Related Questions

What specific projects has David Liu led at AeroVironment?
How did David Liu's experience at Meta influence his role at AeroVironment?
What are some notable achievements of David Liu in his career?
How has David Liu's background in AI and ML impacted his work in robotics?
What challenges has David Liu faced in his engineering leadership roles?
David Liu
David Liu, photo 1
David Liu, photo 2
Add to my network

Location

Los Angeles Metropolitan Area