Intelligent Behavioral Action Aiding For Improved Image Navigation

In egomotion image navigation, errors are common especially when traversing areas with few landmarks. Since image navigation is often used as a passive navigation technique in GNSS (Global Navigation Satellite Systems) denied environments, egomotion accuracy is important for precise navigation in these challenging environments. One of the causes of egomotion errors is inaccurate landmark distance measurement (e. g. sensor noise). This research develops a landmark location egomotion error model that quantifies the effects of landmark locations on egomotion value uncertainty and errors. The error model accounts for increases in landmark uncertainty due to landmark distance and image centrality. A robot then uses the error model to execute actions that position landmarks in image positions that give the least egomotion calculation uncertainty. Three action-aiding solutions are proposed: (1) qualitative non-evaluative aiding action, (2) quantitative evaluative aiding action with physical scans, and (3) quantitative evaluative aiding action with landmark tracking. Simulation results show that both action-aiding techniques reduce the position uncertainty compared to no action aiding. Physical testing results substantiate simulation results. Compared to no action aiding, non-evaluative action aiding reduced egomotion position errors by an average 31.5%, while evaluative action aiding reduced egomotion position errors by an average 72.5%. Physical testing also showed that evaluative action aiding enables egomotion to work reliably in areas with few features, achieving 76% error reduction in egomotion position compared to no aiding.