Current Status and Future Trends on Robot Vision Technology

Hashimoto, Manabu; Domae, Yukiyasu; Kaneko, Shun’ichi

doi:10.20965/jrm.2017.p0275

Cited by 9 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Whether as simple as making the system not work properly, breaking the robot itself, or even injuring and threatening its user's safety. Authors [2] defined some parameters that are important in practice:…”

Section: B Role and Parameters Of Robot Visionmentioning

confidence: 99%

Latest Trends in Visual Manipulation and Navigation in Robotics

Amri

Areche

Naik

2023

JNEST

View full text Add to dashboard Cite

In recent decades, the term Robot has become more and more popular. A robot can be defined as a machine that is specifically built to complete certain tasks to help human-being. In order to successfully accomplish its task, the robot needs to receive input data and process it. Then, the processed data is used for manipulator-actions decision-making. The input data can vary from sound, temperature, vibration, touch, vision, etc. Among those input data, vision is arguably one of the most challenging data. This is because vision often needs detailed and complicated preprocessing before it can be used. In addition, vision data size is relatively larger compared to the other type of input data, making it more challenging to process considering the computational resources. In this paper, current research and future development trend of robotic vision were reviewed and discussed. Further, challenges and potential issues about robot vision, such as safety and privacy concerns, were also discussed.

show abstract

Section: B Role and Parameters Of Robot Visionmentioning

confidence: 99%

Latest Trends in Visual Manipulation and Navigation in Robotics

Amri

Areche

Naik

2023

JNEST

View full text Add to dashboard Cite

show abstract

“…Locating the BROKK camera means recognising it. This is not a trivial task, even state of the art object recognition and learning techniques are only just becoming realistic to use [17]. A simple approach is using fiducials, easily recognised objects placed into the world.…”

Section: B Locating the Brokk Robotmentioning

confidence: 99%

Towards a Cooperative Robotic System for Autonomous Pipe Cutting in Nuclear Decommissioning

Burrell

West

Monk

et al. 2018

2018 UKACC 12th International Conference on Control (CONTROL)

View full text Add to dashboard Cite

A mobile camera is used to support an assisted teleoperation pipe-cutting system for nuclear decommissioning. The base system consists of dual-manipulators with a single mounted Kinect camera. The user selects the object from an onscreen image, whilst the computer control system automatically grasps the pipe with one end-effector and positions the second for cutting. However, the system fails in some cases because of data limitations, for example a partially obscured pipe in a challenging decommissioning scenario (simulated in the laboratory). Hence, the present article develops a new method to increase the use case scenarios via the introduction of mobile cameras e.g. for mounting on a drone. This is a non-trivial problem, with SLAM and ArUco fiducials introduced to locate the cameras, and a novel error correction method proposed for finding the ArUco markers. Preliminary results demonstrate the validity of the approach but improvements will be required for robust autonomous cutting. Hence, to reduce the pipe position estimation errors, suggestions are made for various algorithmic and hardware refinements.

show abstract

“…Suction cups offer a useful gripping solution, particularly in industrial robotics and warehouse applications [1], [2]. Vision-based grasp algorithms, like Dex-Net [3], show promise but struggle to accurately perceive dark or reflective objects, sub-resolution features, and occlusions [4], resulting in suction cup grip failures.…”

Section: Introductionmentioning

confidence: 99%