The ability to estimate self-position accurately and robustly is widely regarded as one of the fundamental preconditions to achieve next-generation car navigation and driver assistance systems.The authors have proposed a localization technology that estimates self-position by matching the three-dimensional map and entire circumference fish-eye cameras' images with a particle filter.It has been confirmed that this technology can estimate self-position with high accuracy and strong robustness if the vehicle moves at low speed on a flat road.In order to estimate self-position on the slopes and during high-speed moving,the authors improve this technology to be able to set the scattering range of particles dynamically by referring to the running condition and gradientin formation of three-dimensional map.