Change detection using weighted features for image-based localization

Autonomous mobile robots are becoming increasingly important in many industrial and domestic environments. Dealing with unforeseen situations is a difficult problem that must be tackled to achieve long-term robot autonomy. In vision-based localization navigation methods, one of the major issues scene dynamics. The autonomous operation may become unreliable if changes occurring dynamic environments not detected managed. Moving chairs, opening closing doors or windows, replacing objects other make conventional methods fail. To deal these challenges, we present novel method for change detection based on weighted local visual features. core idea algorithm distinguish valuable information stable regions from potentially misleading changing. We evaluate framework feature matching by performing series experiments various real-world results show yields an improvement accuracy, compared baseline without detection. addition, experimental evaluation public data set more than 10 000 images reveals proposed outperforms two alternative recorded several months after initial mapping.