Robust Visual Teach-and-Repeat Navigation with Flexible Topo-metric Graph Map Representation

Abstract--Visual T each-and-Repeat Navigation is a direct solution for mobile robot to be deployed in unknown environments. However, robust trajectory repeat navigation still remains challenged due to environmental changing and dynamic objects. In this paper, we propose a novel visual teach-and-repeat navigation system, which consists of a flexible map representation, robust map matching and a map-less local navigation module. During the teaching process, the recorded keyframes are formulated as a topo-metric graph and each node can be further extended to save new observations. T o enhance the place recognition performance during repeating process, instead of using frame-to-frame matching, we firstly implement keyframe clustering to aggregate similar connected keyframes into local map and perform place recognition based on visual frame-to-local map matching strategy. T o promote the local goal persistent tracking performance, a long-term goal management algorithm is constructed, which can avoid the robot getting lost due to environmental changes or obstacle occlusion. T o achieve the goal without map, a local trajectory-control candidate optimization algorithm is proposed. Extensively experiments are conducted on our mobile platform. The results demonstrate that our system is superior to the baselines in terms of robustness and effectiveness. ECENTL Y, mobile robots are widely applied in industrial and household scenes [1]. Visual localization [2], [3] and navigation [4] methods are extensively studied. Under the condition that task route is certain, such as navigating between fixed stations, Visual Teach-and-Repeat (VTR) navigation [5] can avoid fully mapping of the task environment and make deploying robot efficiently. The teaching process is generally controlled by human operator and the robot records visual frames as map along the task route in real-time.