Monocular Visual Teach and Repeat Aided by Local Ground Planarity

Abstract

Visual Teach and Repeat (VT\&R) allows an autonomous vehicle to repeat a previously traversed route without a global positioning system. Existing implementations of VT\&R typically rely on 3D sensors such as stereo cameras for mapping and localization, but many mobile robots are equipped with only 2D monocular vision for tasks such as teleoperated bomb disposal. While simultaneous localization and mapping (SLAM) algorithms exist that can recover 3D structure and motion from monocular images, the scale ambiguity inherent in these methods complicates the estimation and control of lateral path-tracking error, which is essential for achieving high-accuracy path following. In this paper, we propose a monocular vision pipeline that enables kilometre-scale route repetition with centimetre-level accuracy by approximating the ground surface near the vehicle as planar (with some uncertainty) and recovering absolute scale from the known position and orientation of the camera relative to the vehicle. This system provides added value to many existing robots by allowing for high-accuracy autonomous route repetition with a simple software upgrade and no additional sensors. We validate our system over 4.3 km of autonomous navigation and demonstrate accuracy on par with the conventional stereo pipeline, even in highly non-planar terrain.Comment: In Proceedings of the International Conference on Field and Service Robotics (FSR'15), Toronto, Ontario, Canada, Jun. 24-26 201

Robust Monocular Visual Teach and Repeat Aided by Local Ground Planarity and Color‐constant Imagery

“…In our previous conference paper (Clement et al., ), we noted that our monocular VT&R system is less robust to lighting and self‐similar textures than its stereo counterpart. In this work, we address this tradeoff through the use of color‐constant imagery that is robust to shadows and illumination changes in scenes with vegetation, rocks, and sand (Paton et al., ; Ratnasingam & Collins, ).…”

“…Their system has since been extended to other natively three‐dimensional (3D) sensor configurations including appearance‐based lidar (McManus, Furgale, Stenning, & Barfoot, ), multiple stereo cameras (Paton, Pomerleau, & Barfoot, ), and RGB‐D cameras. Recently, Clement, Kelly, and Barfoot () investigated the use of a much simpler sensor configuration—a single 2D monocular camera—in a VT&R system, with 3D information inferred from approximations of local scene geometry and the known position and orientation of the camera relative to the vehicle.…”

“…This paper builds on the work of our recent conference paper (Clement et al., ), in which we proposed to represent the map as a manifold of locally planar ground surfaces with a representation of surface uncertainty, rather than using a full monocular SLAM system to create a globally consistent map. This provides a simple means of generating local metric information from the known height and orientation of the camera on the vehicle, without requiring any specific camera motion.…”

Landing Site Detection for Autonomous Rotor Wing UAVs Using Visual and Structural Information

“…This approach aims to land on an unstructured surface using a position estimate relative to the take-off path of the drone to guide the drone back. The method is inspired by Visual Teach and Repeat [16], where the take-off is the teach pass and the landing is the repeat pass. Fraczek et al [17] presents an embedded Vision system for automated drone landing site detection.…”

Vision-Based Real-Time Positioning and Autonomous Navigation System Using Mobile Robot in Indoor Environments

“…Combining both teach-and-replay feature-based method and a segmentation-based approach, De Cristóforis et al (2015) developed an autonomous navigation method, as an improved version of Chen and Birchfield (2009) method, capable of operating in both indoor and outdoor environment. Following the same logic, teach and replay based technique aided by local ground planarity is used by Clement et al (2015) for an autonomous navigation system experimented in both indoor and outdoor environments. However, as mentioned by De Cristóforis et al (2015), the main drawback of the teach and replay methods lies in the fact that the robot workspace is limited only to the regions mapped during the training step.…”