Through the Looking Glass: Diminishing Occlusions in Robot Vision Systems with Mirror Reflections

Yoshioka, Kentaro; Okuni, Hidenori; Ta, Tuan Thanh; Sai, Akihide

doi:10.1109/iros51168.2021.9636366

Cited by 4 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a result, the PD sensitivity is inevitably lower than that of the dToF method. For these two reasons, it is challenging to achieve the 200 m measurement performance required for ADAS, and iToF may be more suited for short-range applications requiring high precision, such as robotics [37].…”

Section: Basic Lidar Architecturesmentioning

confidence: 99%

A Tutorial and Review of Automobile Direct ToF LiDAR SoCs: Evolution of Next-Generation LiDARs

Yoshioka

2022

IEICE Trans. Electron.

Self Cite

View full text Add to dashboard Cite

LiDAR is a distance sensor that plays a key role in the realization of advanced driver assistance systems (ADAS). In this paper, we present a tutorial and review of automotive direct time of flight (dToF) LiDAR from the aspect of circuit systems. We discuss the breakthrough in ADAS LiDARs through comparison with the first-generation LiDAR systems, which were conventionally high-cost and had an immature performance. We define current high-performance and low-cost LiDARs as nextgeneration LiDAR systems, which have significantly improved the cost and performance by integrating the photodetector, the readout circuit, and the signal processing unit into a single SoC. This paper targets reader who is new to ADAS LiDARs and will cover the basic principles of LiDAR, also comparing with range methods other than dToF. In addition, we discuss the development of this area through the latest research examples such as the 2-chip approach, 2D SPAD array, and 3D integrated LiDARs.

show abstract

Section: Basic Lidar Architecturesmentioning

confidence: 99%

A Tutorial and Review of Automobile Direct ToF LiDAR SoCs: Evolution of Next-Generation LiDARs

Yoshioka

2022

IEICE Trans. Electron.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Occlusion detection has been studied in the literature to improve the performance of perception systems for object tracking. In [15], a simple approach is proposed to detect occluded areas in the sensing data (e.g., point cloud) using a predefined height threshold. When the robot arm is above the target objects that are laid on a table, any point above the threshold is classified as occluded.…”

Section: Introductionmentioning

confidence: 99%

A lightweight method for detecting dynamic target occlusions by the robot body

Sampaziotis¹,

Sotiris²,

Kiatos³

et al. 2023

Preprint

View full text Add to dashboard Cite

Robot vision is greatly affected by occlusions, which poses challenges to autonomous systems. The robot itself may hide targets of interest from the camera, while it moves within the field of view, leading to failures in task execution. For example, if a target of interest is partially occluded by the robot, detecting and grasping it correctly, becomes very challenging. To solve this problem, we propose a computationally lightweight method to determine the areas that the robot occludes. For this purpose, we use the Unified Robot Description Format (URDF) to generate a virtual depth image of the 3D robot model. Using the virtual depth image, we can effectively determine the partially occluded areas to improve the robustness of the information given by the perception system. Due to the real-time capabilities of the method, it can successfully detect occlusions of moving targets by the moving robot. We validate the effectiveness of the method in an experimental setup using a 6-DoF robot arm and an RGB-D camera by detecting and handling occlusions for two tasks: Pose estimation of a moving object for pickup and human tracking for robot handover. The code is available in https://github.com/auth-arl/virtual_depth_image.

show abstract