Qingyang Tan
Papers
2
Total Citations
53
H-Index
2
About
Qingyang Tan is a researcher at the forefront of computer graphics and robotics, specializing in accelerating physical simulations and enabling intelligent multi-agent navigation. His work bridges the gap between high-fidelity physics and real-time performance through deep learning. Tan’s major contributions include a groundbreaking method for real-time simulation of thin-shell deformable materials, where he introduced a CNN-based mesh embedding technique that reduces the high-dimensional configuration space of deformable objects into a low-dimensional feature space, achieving dramatic speedups without sacrificing accuracy. This work, published in 2020, has garnered 31 citations, highlighting its impact on interactive graphics and virtual reality. In parallel, Tan developed DeepMNavigate, a deep reinforcement learning framework for multi-robot navigation that unifies local and global collision avoidance. By processing motion information maps through a three-layer CNN, his algorithm enables robust, decentralized navigation in dense, dynamic environments, earning 22 citations. Tan’s research is notable for its practical applications—from real-time surgical simulation to warehouse robotics—and his ability to integrate geometric deep learning with physical modeling positions him as a rising innovator in embodied AI and computational physics.
Research Focus
Key Achievements
Top Papers
- 1
- 2