Junik Jang

Papers

Total Citations

H-Index

About

Junik Jang is a researcher advancing the field of autonomous mobile robotics, with a focus on deep learning and efficient visual perception. His key contributions lie in developing end-to-end convolutional neural network (CNN) frameworks for robot navigation, as demonstrated in his highly cited 2018 work, which has garnered 53 citations. This paper pioneered a streamlined approach that bypasses traditional multi-step navigation pipelines, enabling robots to learn directly from raw camera inputs to control actions—a significant step toward more adaptive and intelligent autonomous systems. Jang also addresses the critical challenge of computational efficiency in his 2018 study on light-weight visual place recognition, which has earned 6 citations. This work targets embedded systems, proposing a compact CNN architecture that balances accuracy with reduced computational load, making it suitable for resource-constrained mobile robots. By tackling both high-performance navigation and practical deployment constraints, Jang’s research bridges the gap between cutting-edge deep learning and real-world robotic applications, offering valuable insights for students and engineers working on autonomous systems, computer vision, and embedded AI.