A ToF-Camera as a 3D Vision Sensor for Autonomous Mobile Robotics

Abstract

The aim of this paper is to deploy a time-of-flight (ToF) based photonic mixer device (PMD) camera on an Autonomous Ground Vehicle (AGV) whose overall target is to traverse from one point to another in hazardous and hostile environments employing obstacle avoidance without human intervention. The hypothesized approach of applying a ToF Camera for an AGV is a suitable approach to autonomous robotics because, as the ToF camera can provide three-dimensional (3D) information at a low computational cost, it is utilized to extract information about obstacles after their calibration and ground testing, and is mounted and integrated with the Pioneer mobile robot. The workspace is a two-dimensional (2D) world map which has been divided into a grid/cells, where the collision-free path defined by the graph search algorithm is a sequence of cells the AGV can traverse to reach the target. PMD depth data is used to populate traversable areas and obstacles by representing a grid/cells of suitable size. These camera data are converted into Cartesian coordinates for entry into a workspace grid map. A more optimal camera mounting angle is needed and adopted by analysing the camera's performance discrepancy, such as pixel detection, the detection rate and the maximum perceived distances, and infrared (IR) scattering with respect to the ground surface. This mounting angle is recommended to be half the vertical field-of-view (FoV) of the PMD camera. A series of still and moving tests are conducted on the AGV to verify correct sensor operations, which show that the postulated application of the ToF camera in the AGV is not straightforward. Later, to stabilize the moving PMD camera and to detect obstacles, a tracking feature detection algorithm and the scene flow technique are implemented to perform a real-time experiment.

Recent Advances in 3D Data Acquisition and Processing by Time-of-Flight Camera

“…TOF camera provides good obstacle avoidance information for automatic driving [57]- [60]. In industrial production, cameras guide the robots on the installation, quality control and raw material selection [61], [62].…”

“…TOF cameras as a developing type of 3D vision sensor are attracting more and more attentions for autonomous mobile robotics [42], [57], [201]. The real-time 3D data can help robots to accomplish autonomous path-planning [57], [202]. Therefore, the vision system now becomes indispensable for the robots to see the environment and avoid possible obstacles, e.g.…”

Mobil robot mozgásának vezérlése

“…Egy ilyen jármű a Miskolci Egyetem Logisztikai Intézetében levő High-Tech laboratórium anyagmozgatási rendszerének része [4]. A pozíció meghatározására különböző technológiák állnak rendelkezésre [5], mint például induktív szenzor [6][7][8], képfeldolgozás [9,10], vagy LIDAR. A vizsgált jármű pozícióját egy LIDAR szenzor segítségével érzékeli és virtuális pályát használ, mint a [11]-ben kifejlesztett AGV.…”

Depth Map Enhancement based on Improved Adaptive Total Generalized Variation for ToF Camera

“…In equation (6) So the magnitude and the phase are given for each scale by the equation (7) and the equation (8).…”

“…The disadvantages are obvious such as low resolution, motion artifacts, edge distortion, ambient light noise, et al, [1,5]. Despite its disadvantages, it is already showing great potential in many applications where fast 3D depth data is needed, such as pose estimation [6,7], obstacle avoidance [8,9] and others.…”