Considering the terminal guidance problem with impact angle constraint and overload constraint,an Adaptive Fuzzy-Sliding Mode Control (AFSMC) guidance law is presented.Its stability is proved by Layapunov stability criterion.The desired impact angle is defined by a desired Line of Sight (LOS) angle,and achieved by design of sliding mode surface.The Adaptive Neuro Fuzzy Inference System (ANFIS) is introduced to reduce the high-frequency chattering of sliding mode control.Since the ANFIS has a powerful self-learning ability,it is applied to adaptive update the additional acceleration command.As a result,the guidance law has good robustness to uncertain chattering and model uncertainty,and the guidance accuracy is improved.Simulation results show that the AFSMC guidance law can meet the requirements of impact angle constraint,overload constraint and miss distance constraint in various scenarios.