Guizhi Li

Beijing Information Science & Technology University, University of Science and Technology Beijing

Papers

Total Citations

H-Index

About

Guizhi Li’s research lies at the intersection of computer vision, mobile robotics, and adaptive learning systems, with a focus on enabling intelligent agents to perceive and navigate complex environments. A central contribution is the development of a color image segmentation method based on rival penalized competitive learning (RPCL), which autonomously determines the optimal number of color clusters—a critical step for tasks like object recognition and scene understanding. Li also pioneered the concept of “Softman,” a virtual robot designed to address information complexity, security vulnerabilities, and the lack of humanized services on the Internet, proposing a novel path for network intelligence. In mobile robotics, Li advanced reinforcement learning with adaptive state space construction, improving generalization and learning efficiency for navigation in unknown environments. Another notable achievement is the doorplate adaptive detection and recognition system, which uses visual landmarks for robust self-localization of indoor robots, reducing reliance on memory-intensive environmental maps. With key papers accumulating citations in the single digits, Li’s work is recognized for its foundational ideas in adaptive segmentation and robot autonomy, contributing to the evolution of intelligent systems that learn and adapt in real-world settings.