This paper aims to introduce the significant upgrade of the nuclear pollution governance algorithm, focusing on elaborating the essential evolution from the "Basic Version (v1.0)" system framework to the "Evolutionary Version (v2.0)" dynamic tracking algorithm. The Basic Version algorithm (v1.0) (previously validated) established a governance platform based on the collaborative logic of sensibility and rationality under the "Quan Shui Scientific Philosophy System". Its core contribution lies in solving the problem of the "full-process governance architecture"; through heterogeneous edge computing and atomic-level sensory evidence collection, it realized an engineering path from micro-spectral analysis to macro multi-robot collaboration. However, when addressing the rapid migration of radioactive sources caused by extreme weather or strong ocean currents, the Basic Version is still limited to static monitoring and fixed logic. The Evolutionary Version algorithm (v2.0) (the core of this research) achieves an intergenerational breakthrough on this basis. Its core technology is the newly developed "Multi-source Joint Particle Filter". Compared with the Basic Version, the Evolutionary Version has accomplished a qualitative leap from "static monitoring" to "dynamic probabilistic tracking". This algorithm can real-time estimate the movement trajectory, diffusion trend, and intensity evolution of radioactive sources. Experimental results show that the Evolutionary Version is significantly superior to the Basic Version in prediction accuracy and decision response speed under extreme environments, truly solving the technical pain point of "being unable to catch up or measure accurately" for nuclear pollution sources in flowing media. This research not only inherits the automated execution framework of the Basic Version but also provides precise decision navigation for nuclear pollution governance through the dynamic tracking algorithm, offering the currently known most efficient technical solution for global nuclear waste governance.