Aimed at the problem that the detection performance losses of multi-input multiple-output radar due to the inaccurate prior knowledge of the distribution characteristics of target impulse response and clutter impulse response, a joint optimization algorithm of constant modulus robust waveform and receiver filter was proposed. The optimization problem in which the prior knowledge of the distribution characteristics of target impulse response and clutter impulse response is not accurate was modeled as a maximum minimization problem. During the optimizations, the joint design problem was divided into two steps: the optimization problem of receiver filter weights when the waveform was fixed was modeled as the generalized Rayleigh quotient model, and the corresponding weight vector of the receiver filter was obtained; the waveform optimization problem of fixed weight was solved by using positive semi-definite relaxation technique to obtain the corresponding waveform matrix, and the constant modulus waveform was obtained by Gauss random method according to the waveform matrix obtained. The convergence of the proposed algorithm was proved, and the simulation results show the effectiveness of the proposed algorithm. 