Robot navigation and obstacle detection in pipelines using time-of-flight imagery

Abstract

Range imagery provided by time-of-flight (TOF) cameras has been shown to be useful to facilitate robot navigation in several applications. Visual navigation for autonomous pipeline inspection robots is a special case of such a task, where the cramped operating environment influences the range measurements in a detrimental way. Inherent in the imaging system are also several defects that will lead to a smearing of range measurements. This paper sketches an approach for using TOF cameras as a visual navigation aid in pipelines, and addresses the challenges concerning the inherent defects in the imaging system and the impact of the operating environment. New results on our previously proposed strategy for detecting and tracking possible landmarks and obstacles in pipelines are presented. We consider an explicit model for correcting lens distortions, and use this to explain why the cylindrical pipe is perceived as a cone. A simplified model, which implicitly handles the combined effects of the environment and the camera on the measured ranges by adjusting for the conical shape, is used to map the robot's environment into an along-axis-view relative to the pipe, which facilitates obstacle traversal. Experiments using a model pipeline and a prototype camera rig are presented.

Active vision in robotic systems: A survey of recent developments

“…Its task is to incrementally build up a representation of its surroundings (Suppa and Hirzinger 2007;. A local navigation strategy has to be implemented for unknown environment exploration (Amin et al 2008;Radovnikovich et al 2010;Thielemann et al 2010).…”

Localization, Mapping, Navigation, and Inspection Methods in In-Pipe Robots: A Review

“…The landmark detection method in this paper is based on the outside of the pipe and voids that existed on the soil. The authors in [73] took advantage of the 3D data provided by the TOF camera to find significant deviations from the cylindrical shape of the pipeline. To facilitate the obstacle traversal and find the obstacle position relative to the pipeline, a simplified model is used to map the robot's environment into an along-axis view.…”

“…In [69] and [81], the authors proposed an HMM to localize a robot in the pipeline. In [73], the hidden state has three components, including the robot's current position (𝑋 𝑡 ), previous position (𝑋 𝑡−1 ) and current orientation (𝜃 𝑡 ). The robot state is only updated at junctions or the ends of pipes.…”