To effectively address the intricate airborne multi-sensor multi-task allocation challenge, a self-adaptive guided differential evolution algorithm based on niching meta-knowledge transfer (NMT-AGDE) is proposed. This novel approach fosters population evolution via a trifecta of mechanisms: small habitat population clustering, meta-knowledge migration, and parameter tuning coordination. Furthermore, we propose a dynamic reconfiguration algorithm for the task allocation decision chain, which is grounded in an unexpected event response mechanism. This algorithm enables us to design a local release mechanism and reconfigure the decision chain, thereby overcoming issues such as inadequate detection capability and sluggish response speeds in dynamic environments where sensors encounter unexpected events. Through rigorous simulation and comparative experiments, we demonstrate that our NMT-AGDE-based multi-sensor cooperative task allocation algorithm surpasses traditional intelligent optimization algorithms in terms of convergence speed and solution quality. Notably, it exhibits rapid emergency response capabilities with minimal computational overhead.