Yiyang Wang

Dalian Maritime University, Tsinghua University

Papers

Total Citations

H-Index

About

Yiyang Wang is a researcher advancing the frontiers of autonomous systems through innovative work in pedestrian trajectory prediction and real-time visual SLAM (Simultaneous Localization and Mapping). Their key research areas span trajectory forecasting, hardware acceleration for robotic perception, and energy-efficient embedded systems. Wang’s most notable contribution is the development of SEEM (Sequence Entropy Energy-Based Model), a novel framework for pedestrian trajectory prediction that addresses critical limitations in diversity, accuracy, and stability—garnering 40 citations since 2022. This work has direct implications for autonomous driving and social robotics, where reliable human motion forecasting is essential. In parallel, Wang has made significant strides in hardware-optimized SLAM systems, designing a 325 FPS corner-detection accelerator and an energy-efficient pose-estimation FPGA accelerator for real-time mobile V-SLAM robots. These contributions tackle the computational bottlenecks of visual odometry, enabling robust, low-power robotic navigation. By bridging algorithmic innovation with hardware efficiency, Wang’s research demonstrates a rare combination of theoretical depth and practical deployment—making their work highly relevant for students and engineers building next-generation autonomous platforms.