Song Zeng

Papers

Total Citations

H-Index

About

Song Zeng is a robotics researcher whose work centers on bio-inspired locomotion and autonomous navigation in unstructured environments. His most influential contribution, "An Inchworm-inspired Crawling Robot" (2019, 29 citations), demonstrates a novel approach to soft robotics by mimicking the peristaltic motion of inchworms, achieving efficient and adaptable crawling on varied surfaces. This work has become a foundational reference for researchers exploring biomimetic gait design and compliant mechanisms. Zeng has also advanced the field of outdoor terrain recognition through his 2021 study, which employs transfer learning to enable mobile robots to classify terrain types—such as gravel, grass, or asphalt—without relying on traditional color or texture features. By leveraging pre-trained neural networks, his method significantly improves generalization across different environments, directly impacting gait planning and speed control for field robots. These contributions highlight Zeng’s dual focus on mechanical innovation and intelligent perception, bridging the gap between biological inspiration and practical robotic deployment. His work continues to influence the development of more resilient and autonomous robots capable of navigating complex, real-world terrains.