Detail of the student project

Topic:Path adaptation for information gain collection
Department:Katedra kybernetiky
Supervisor:Kelen Cristiane Teixeira Vivaldini, Ph.D.
Announce as:Diplomová práce, Bakalářská práce, Semestrální projekt
Description:Unmanned aerial vehicles (UAVs) offer a relatively risk-free and low-cost way to quickly and systematically observe phenomena and targets at high spatio-temporal resolution. For identifying and monitoring targets in unknown areas, path planning can be used to cover the entire area and adapt the planning to gain information during the execution of the trajectory. That is, when the UAV identifies a target, it can decide between decreasing its speed and/or altitude to obtain more information, increasing the information gain for the mission, or using the previously acquired knowledge to make the decision. Thus, when it finds a pattern among the detected targets, it can decide not to carry out a more specific recognition validation. Thus, in this work, we will approach informative trajectory planning techniques and machine learning together, which is promising for the area of decision making in path planning.
Bibliography:- P. Zhong, B. Chen, S. Lu, X. Meng and Y. Liang, "Information-Driven Fast Marching Autonomous Exploration With Aerial Robots," in IEEE Robotics and Automation Letters, vol. 7, no. 2, pp. 810-817, April 2022, doi: 10.1109/LRA.2021.3131754.
- Hitz, Gregory, et al. "Adaptive continuous‐space informative path planning for online environmental monitoring." Journal of Field Robotics 34.8 (2017): 1427-1449.
- S. Zhang, R. Cui, W. Yan and Y. Li, "Informative Path Planning for AUV-based Underwater Terrain Exploration with a POMDP," 2021 China Automation Congress (CAC), Beijing, China, 2021, pp. 4756-4761, doi: 10.1109/CAC53003.2021.9728147.
Responsible person: Petr Pošík