Grammar-based parsing is a prevalent method for natural language under-standing (NLU) and has been introduced into dialogue systems for spoken language processing(SLP). A robust parsing scheme is proposed in this paper to overcome the notorious phenome-na, such as garbage, ellipsis, word disordering, fragment, and ill-form, which frequently occurin spoken utterances. Keyword categories are used as terminal symbols, and the definition ofgrammar is extended by introducing three new rule types, by-passing, up-messing and over-crossing, in addition to the general rules called up-tying in this paper, and the use of semanticitems simplifies the semantics extraction. The corresponding parser marionette, which is essen-tially a partial chart parser, is enhanced to parse the semantic grammar. The robust parsingscheme integrating the above methods has been adopted in an air traveling information ser-vice system called EasyFlight, and has achieved a high performance when used for parsingspontaneous speeches.