Papers
4
Total Citations
55
H-Index
3
About
Di Deng’s research lies at the intersection of autonomous robotics, path planning, and multi-agent exploration, with a particular focus on enabling robots to intelligently navigate and map unknown environments. His major contribution is the development of a frontier-based automatic-differentiable information gain measure, a novel heuristic that bridges the gap between traditional frontier-based and information-theoretic exploration methods. This work, published in 2020 and cited 33 times, allows robots to more efficiently evaluate the value of potential viewpoints by making the information gain metric differentiable, thereby enabling gradient-based optimization for path planning. Deng extended this concept to 3D environments (9 citations) and further advanced the field by proposing a coordinated aerial-ground robot exploration framework using Monte-Carlo view quality rendering. His earlier work on sensor-guided robot path generation for surface repair tasks on large-scale buoyancy modules (11 citations) demonstrates the practical application of his algorithms in industrial settings, such as offshore oil and gas maintenance. Through these contributions, Deng has significantly improved the scalability and efficiency of autonomous exploration in complex, unstructured environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4