The invention discloses a weighted subspace data fusion direct positioning method based on an augmented co-prime array, and the method specifically comprises the following steps: Z01, obtaining
radiation source signals of an unmanned plane at different position observation stations, and carrying out the sampling
processing of the signals; z02, calculating a
covariance matrix according to the array output
signal obtained after sampling
processing, and obtaining a characteristic value according to the
covariance matrix; z03, obtaining a
noise subspace according to the
covariance matrix, and enabling the position of each
observation point to obtain an
optimal weight through the minimum total projection error at the observation points; and Z04, obtaining a
loss function of a positioning
algorithm according to mutual orthogonality of the
signal manifold vector and the
noise subspace, and obtaining an
estimation result of the position of the
radiation source through a search formula. According to the method, an orthogonal projection error is balanced, a
loss function with small error and high robustness is obtained, and the positioning precision is improved; and compared with a uniform array
algorithm, the degree of freedom is obviously improved, the spatial degree of freedom of the array is expanded, and the number of identification sources is increased.