This paper considers an extremum-seeking-based source seeking with dynamic sensor reading and a class of nonlinear maps to localise a stationary source emitting from an unknown signal field. Unlike the majority of existing extremum-seeking-based source seeking strategies that only deal with the quadratic map, the more universal case is considered that the map is quasi-concave nonlinear. In addition, the dynamic sensor reading is also considered which arises in many applications. However, this dynamic sensor could cause poor performance and could be a potential cause of instability. To resolve it, combined filters and a phase-lead compensator are utilised to improve the performance. Furthermore, a critical semi-global stability analysis is theoretically proven which shows that the proposed scheme can seek an arbitrarily small neighbourhood of the signal's source from an arbitrarily large set of initial location. Finally, the effectiveness of the proposed scheme is illustrated by numerical simulations.

本文考虑使用动态传感器读数和一类非线性映射来定位从未知信号场发射的固定源的基于极值搜索的源搜索。与大多数现有的仅处理二次图的基于极值搜索的寻源策略不同，更普遍的情况是认为该图是准凹非线性的。此外，还考虑了在许多应用中出现的动态传感器读数。但是，这种动态传感器可能会导致性能不佳，并且可能是不稳定的潜在原因。为了解决这个问题，使用组合滤波器和相位超前补偿器来提高性能。此外，理论上证明了关键的半全局稳定性分析，表明所提出的方案可以从任意大的初始位置集合中寻找信号源的任意小邻域。最后，通过数值模拟说明了所提方案的有效性。