Abstract: Aiming at the problem of limited training samples and target information inaccuracy and other sensitive factors leading to a sharp decline in detection performance when detecting targets by multi-channel array radar, this paper proposes a covariance structure-assisted adaptive target detection method with full incremental linearization model. The method adopts the idea of joint processing, modeling the target inaccurate information as a full incremental linearized model by array oriented vectors, and then designing the covariance structure-assisted detection by using unitary matrix transformation, and transforming this detection problem into a fractional optimization problem, and then deducing the final detection statistics by whitening and optimizing the solution. Numerical simulation results show that by assisting the optimization of the full incremental linearization model using the covariance structure information, the detection performance of the target in the complex and sensitive environments is effectively improved, and compared with traditional detection methods, the detection performance remains optimal when the adaptive sample size is reduced under specific parameter conditions.