Optimal Spatial-Temporal Triangulation for Bearing-Only Cooperative Motion Estimation

• DOI: 10.48550/arxiv.2310.15846
• Publication Date: 2023-01-01
• Authors: Chao Zheng; Yongqing Mi; Hao Guo; H. B. Chen; Zheng Yan Lin; S. Y. Zhao
• Keywords: FOS: Computer and information sciences; Computer Science - Robotics; Robotics (cs.RO)

Vision-based cooperative motion estimation is an important problem for many multi-robot systems such as aerial target pursuit. This can be formulated bearing-only estimation, where the visual measurement modeled a bearing vector pointing from camera to target. The conventional approaches are mainly based on framework distributed Kalman filtering (DKF). In this paper, we propose new optimal algorithm, named spatial-temporal triangulation, method of recursive least squares, which provides more flexible designing estimators than DKF. design algorithm fully incorporates all available information and specific triangulation geometric constraint. As result, has superior performance state-of-the-art DKF algorithms in terms both accuracy convergence speed verified by numerical simulation. We rigorously prove exponential proposed algorithm. Moreover, verify effectiveness under practical challenging conditions, develop vision-based pursuit system, first autonomous so far best our knowledge.