Papers
4
Total Citations
19
H-Index
3
About
Hui Xiong is a researcher working at the intersection of autonomous robotics, edge computing, and multimodal perception. His work addresses two principal challenges in intelligent systems: efficient computational resource management and advanced environmental understanding for autonomous agents. In the domain of computation and task scheduling, Xiong has made notable contributions by developing deep reinforcement learning frameworks — including both graph neural network (GNN)-based and hierarchical approaches — to optimize task offloading and parallel scheduling within autonomous multi-robot systems (AMRS). These methods tackle the critical problem of distributing compute-intensive workloads in environments lacking external infrastructure, improving system responsiveness and adaptability. His event-driven multimodal prediction approach further advances scheduling by dynamically accounting for shifting task priorities. Complementing this, Xiong has pioneered work in multimodal embodied perception, most notably through Talk2Radar, which bridges natural language processing with 4D millimeter-wave radar for 3D referring expression comprehension — extending intelligent perception beyond vision-centric paradigms. With his most-cited works accumulating up to six citations in 2025, Xiong represents an emerging voice in robotics and AI, contributing foundational methods that push autonomous systems toward greater intelligence and real-world resilience.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4