Chaoqi Liu

I'm a sophomore at University of Illinois Urbana-Champaign (UIUC) studying computer science and mathematics. My academic journey is driven by an unwavering ardor for the field of robotics, especially mobile manipulation. Currently, I'm working on large-scale Graph-based Neural Dynamics (GBND), under the esteemed guidance of Prof. Kris Hauser and Prof. Yunzhu Li. Previously, I enjoyed exploring model-based planning and control, and had a great time collaborating with Ramkumar Natarajan and Prof. Maxim Likhachev on kinodynamic motion planning by graph search and optimization.

chaoqil2 [at] illinois [dot] edu / GitHub / Twitter

Publications

Ramkumar Natarajan, Chaoqi Liu, Howie Choset, Maxim Likhachev
Implicit Graph Search for Planning on Graphs of Convex Sets
RSS, 2024

A scalable and efficient way to plan on Graphs of Convex Sets (GCS) with stronger theoretical properties. We plan on GCS using a previously developed hybrid search-optimization framework called INSAT.

Abstract

Smooth, collision-free motion planning is a fundamental challenge in robotics with a wide range of applications across various industries and domains such as automated manufacturing, search & rescue, underwater exploration, etc. Graphs of Convex Sets (GCS) is a recent method for synthesizing smooth trajectories by decomposing the planning space into convex sets, forming a graph to encode the adjacency relationships within the decomposition, and then simultaneously searching this graph and optimizing parts of the trajectory to obtain the final trajectory. To do this, one must solve a Mixed Integer Convex Program (MICP) and to mitigate computational time, GCS proposes a convex relaxation that is very tight. Despite this tight relaxation, motion planning with GCS for real-world robotics problems translates to solving the simultaneous batch optimization problem that may contain millions of constraints and therefore can be slow. This is further exacerbated by the fact that the size of the GCS problem is invariant to the planning query. Based on the observation that the trajectory solution lies only on a fraction of the set of convex sets, we present two implicit graph search methods for planning on the graph of convex sets called INSATxGCS (IxG) and IxG*. INterleaved Search And Trajectory optimization (INSAT) is a previously developed algorithm that alternates between searching on a graph and optimizing partial paths to find a smooth trajectory. We show that by using an implicit graph search method INSAT on the graph of convex sets, we can plan faster and provide stronger guarantees on completeness and optimality. In addition, introducing a search-based technique to plan on the graph of convex sets enables us to easily leverage well-established techniques such as search parallelization, lazy planning, anytime planning, and replanning as future work. We compare our method with GCS to show numerically that IxG outperforms in several applications including planning for an 18-DoF multi-arm assembly scenario.

Projects

	Multi-agent Path Finding (MAPF) on Grids CS598KKH Advanced Computational Topics in Robotics @ UIUC, 2023fa report This is a course project at UIUC. Explored how multiple robots can navigate in a grid, such as those that navigate automated warehouses. Analyzed the completeness and time complexity, and implemented a naive algorithm provided in the project prompt. Designed and implemented its simultaneous variant. Discussed ways to improve the performace, e.g., introducing some heuristics, and Collision Based Search (CBS) framework.
	LALip: Large Language Model Aided Lip-Reading CS543 Computer Vision @ UIUC, 2023fa code / report This is a course project at UIUC. Designed and implemented LALip, integrating deep-learning computer vision models with a large language model (LLM) to expedite training and enhance accuracy. Implemented methodology for lip frame extraction, grayscale conversion, and lip region isolation using GRID corpus data.
	Spimbot: A bot that Aced the Arena CS233 Computer Architecture @ UIUC, 2023sp code This is a course project at UIUC. Wrote a 2000+ lines MIPS Assembly code program to manipulate and control a bot that solves puzzles and cleans tiles. Utilized code optimization techniques such as double buffering, loop unrolling, parallelism. Resulting in solving puzzles up to 7 times faster than the provided solver. Bot developed had the best locomotion and fastest puzzle solving ability among all bots. Aced the competition and ranked 1st among over 150+ teams.
	Huffman Coding CS128 Introduction to Computer Science II @ UIUC, 2022fa code / demo This is a course project at UIUC. Implemented Huffman Coding algorithm with Rust, provides basic encoding/decoding of text content, as well as image compression. Detailed explanations and technical details are shown in the README.md of the repository.

Teaching and Service

Course Assistant, CS233 Computer Architecture @ UIUC, 2023fa

Honor and Awards

UIUC Dean's List, 2022fa, 2023sp, 2023fa

Misc.

I have consistently maintained a perfect 4.0 GPA, excelling with top grades of A/A+ in all of my Computer Science, Mathematics, and Statistics courses. A complete list can be found here.

刘晁企 is my name in Chinese. You can pronounce it as "Chow-Chee" for my first name and "Lee-oh" for my last name.

Page template borrowed from Jon Barron