Alfred Cueva

I am an incoming Master's student at Georgia Institue of Technology, majoring in Robotics. Currently I work as a Robotics Engineer at Samsung C&T on system integration for collaborative robots.

I received my Bachelors in Mechanical Engineering from Seoul National University during which I was a Research Intern at Dynamic Robotics System Lab, advised by Prof. Jaeheung Park and Dr. Daegyu Lim focusing on Reinforcement Learning and Control for legged locomotion. I also interned at Soft Robotics & Bionics Lab under the guidance of Prof. Yong-Lae Park on soft robotic sensing for industrial robots.

I co-organized a non-profit organization dedicated to AI education, AI Tech Play, and hosted the first nationwide AI camp focused on autonomous racing competitions for high school students.

Fun Fact: I am fluent in 4 languages.

Email / CV / Linkedin / Github

Research

I am particularly interested in developing control frameworks that enable robots adapt to diverse tasks and environments by combining learning-based methods with classical control. My dream is to see robots perform complex, long-horizon tasks in real environments, just like humans do. I am really eager to explore agile lomocotion, loco-manipulation and multi-agent coordination.

RL-Policy Guided Optimal Design of Parallel Elastic Actuator for Weak Actuation of Bipedal Robot
Alfred Cueva, Jaeheung Park, Yong-Lae Park
Bachelor's Thesis (Outstanding BS Thesis Presentation Award)

We design an optimization framework using model-free meta reinforcement learning, and its application to optimizing energy consumption and actuator parameters of bipedal robots with weak actuation.

Projects

	Autonomous Drilling Robot for Cluttered environmentst Project at Samsung C&T (Smart Construction Robotics Challenge Winner) A novel autonomous drilling robot utilizing rule-based computer vision to enable precise, human-like surface drilling on construction sites.
	Reinforcement Learning Agent for Rapid Task Adaptation Alfred Cueva, Gene Chung, Taehung Kim, Sumin Ye Graduate Course Project (Reinforcement Learning, Spring 2023) We combine off-policy reinforcement learning with Seq2Seq networks to perform online rapid adaptation in unknown environments.
	Constrained 2D Online Bin Packing Problem using Reinforcement Learning We learn 2D optimal bin packing strategy with Heuristics Integrated Deep Reinforcement Learning as in Yang et al. , and perform model optimization.
	Control Techniques for Humanoid Robots Alfred Cueva* Graduate Course Project (Theory and Practice of Humanoid Walking Control, Fall 2022) We implement various control algorithms from scratch on Tocabi Robot.
	RC Car Autonomous Driving Alfred Cueva, Max Acosta Deployed path tracking and planning algorithms for autonomous RC cars in environments with various difficulties.
	Soil Sensing with Machine Learning and Satellite Imagery Alfred Cueva, Andy Kim We leverage machine learning regression models to estimate soil health at scale from satellite imagery, supporting data-driven, pro-soil policy enforcement.
	Video Generation from Single-Image Input Alfred Cueva* We propose a novel video-generation algorithm that generates a video from just one image by using sequential structure, without exhibiting any awkward flow problem.

Template from Jon Barron.