Jixuan Zhi

Papers

Total Citations

H-Index

About

Jixuan Zhi is a robotics and artificial intelligence researcher whose work bridges autonomous multi-agent systems, deep reinforcement learning, and human-robot interaction. His most recognized contribution lies in tackling the robotic shepherding problem — the challenge of controlling and navigating coherent groups of agents through complex environments using a single external robot. His 2021 paper on training robust shepherding behaviors using deep reinforcement learning garnered 33 citations, demonstrating that machine learning approaches could overcome the significant limitations of prior obstacle-free methods. Building on this foundation, his 2022 follow-up work extended these capabilities to larger agent groups in obstacle-filled environments through an optimized surrogate framework, pushing the boundaries of scalable multi-agent control. Beyond swarm dynamics, Zhi has made meaningful contributions to the design of human-robot coexistence spaces, recognizing that the physical environments surrounding human-robot interactions profoundly affect safety, comfort, and operational efficiency. This interdisciplinary perspective — spanning reinforcement learning, robotics, and spatial design — positions Zhi as a thoughtful contributor to the increasingly critical question of how intelligent robots can operate safely and effectively alongside both people and other agents in real-world, cluttered environments.