Loading...

Decentralized Relative Position and Heading Estimation in Formation Flight Considering Geometric Constraints

Alizad Arabloue Bisheh, Meysam | 2025

60 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 58495 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Nobahari, Hadi
  7. Abstract:
  8. This thesis addresses the problem of decentralized relative localization in multi-UAV systems operating in GNSS-denied environments. A novel algorithm is developed that leverages relative range and inertial measurements to jointly estimate relative positions, velocities, heading angles, and acceleration biases. By incorporating geometric constraints—such as the law of cosines and velocity relations—into the estimation process, the proposed approach significantly reduces bias errors and improves overall accuracy. Each group of three UAVs forms a triangular configuration, enabling members to exploit spatial relations and share local estimates. A master–regular structure is also considered, where the master UAV estimates and distributes bias values to enhance consistency across the network. An observability analysis establishes less restrictive conditions compared to previous methods. The effectiveness of the algorithm is demonstrated through simulations under noise, initialization uncertainties, and communication disconnections, as well as through experimental validation. Comparative results with consensus-based Kalman filtering and other existing approaches show that the proposed method achieves superior accuracy and holds potential for enhancing absolute navigation in GNSS-denied environments
  9. Keywords:
  10. Location Estimation ; Multi-Agent Motion Planning ; Decentralized Estimation ; Formation Flight ; Relative Localization ; Geometric Constraints

 Digital Object List

 Bookmark

No TOC