Kyunghoon Cho
Papers
7
Total Citations
34
H-Index
4
About
Kyunghoon Cho is a researcher at the forefront of robotic path planning, specializing in the integration of formal mission specifications with safe, autonomous navigation. His work centers on combining Linear Temporal Logic (LTL) with sampling-based and deep learning methods to ensure robots can complete complex tasks while accounting for real-world uncertainties. Cho’s major contributions include developing robust, multi-layered sampling-based algorithms that generate provably safe paths under chance constraints, as well as pioneering end-to-end deep learning frameworks—such as the Transformer Variational Autoencoder—that seamlessly embed LTL specifications into trajectory optimization. His most cited paper, “Chance-Constrained Multilayered Sampling-Based Path Planning for Temporal Logic-Based Missions” (2020), has garnered 15 citations and addresses the critical challenge of mission failure due to noise and obstacles. Cho has also advanced hierarchical learning for autonomous driving, where rule specifications guide safe vehicle control. By bridging formal logic with modern AI, his work is shaping the next generation of reliable, mission-aware robots for applications ranging from environmental monitoring to autonomous vehicles.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5End-to-End Path Planning Under Linear Temporal Logic Specifications2 citations · 2024
- 6Efficient graph-based informative path planning using cross entropy2 citations · 2016
- 7