Design of a Deep Post Gripping Perception Framework for Industrial Robots

Zoghlami, Firas; Kurrek, Philip; Jocas, Mark; Masala, Giovanni Luca; Salehi, Vahid

doi:10.1115/1.4048204

Cited by 3 publications

(3 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Most of the research is focused on applications that have publicly available large data-sets on commonly used objects, whereas in industrial applications object types can be specific to the respective product. Model-free grasping techniques partly addresses data absence [17] and can proceed without having prior knowledge of the objects, but this method complicates the post-gripping [33] process and can introduce additional steps for precise positioning. The latest achievements in the field of computer-generated imagery widen opportunities in synthetic real-life like data generation for object detection tasks in this particular scenario.…”

Section: Related Workmentioning

confidence: 99%

Synthetic Data of Randomly Piled, Similar Objects for Deep Learning-Based Object Detection

Arents,

Lesser,

Bizuns

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Currently, the best object detection results are achieved by supervised deep learning methods, however, these methods depend on annotated training data. With the synthetic data generation approach, we intend to mimic the real data characteristics and diversify the dataset by a systematic rendering of highly realistic synthetic pictures. We systematically explore how different combinations and portions of real and synthetic datasets affect object detectors performance. The developed synthetic data generation framework shows promising results in deep learning-based object detection tasks and can supplement real data when the variety of real training data is insufficient. However, when synthetic data ratio increases over real data ratio, a decrease in average precision can be observed, which has the most affect on 0.75-0.95 IoU threshold range.

show abstract

Section: Related Workmentioning

confidence: 99%

Synthetic Data of Randomly Piled, Similar Objects for Deep Learning-Based Object Detection

Arents,

Lesser,

Bizuns

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…In [10,15,18,21,22,25,[29][30][31]35,37], work is widely carried out under the classification of gripping devices of industrial robots. The methods of recognition [19,23], grasping [13,18,32,33], and manipulation [8,11,12,34] for different gripping devices vary greatly and depend on the object. Among all the gripping devices of industrial robots, pneumatic jet grippers have the greatest dependence on the input parameters of their power characteristics, which are not specified in their technical characteristics.…”

Section: Introductionmentioning

confidence: 99%

Influence of Inlet Parameters on Power Characteristics of Bernoulli Gripping Devices for Industrial Robots

Mykhailyshyn

Xiao

2022

Applied Sciences

View full text Add to dashboard Cite

There is a wide variety of gripping devices with various parameters at the present stage of development of robotics. However, the existing literature about the power characteristics of pneumatic gripping devices does not provide any analysis of the different types of input parameters other than supply pressure and flow characteristics, whereas the input parameters completely determine the technical characteristics of the gripping devices of industrial robots. In particular, for pneumatic grippers, the input parameters play a crucial role in their productivity and energy efficiency. This paper fills the gap by providing a study of the impact of such parameters on the power and energy characteristics of Bernoulli ejection grippers for industrial robots, including: the parameters of the compressed air inlet design, the parameters of the fitting for compressed air supply, the diameter of the hose for compressed air supply, and the size of the vacuum zone. First, this paper presents the results of theoretical studies and finite element method for determining the distribution of pressure on the surface of an object of manipulation, which allows one to determine the lifting force of the Bernoulli gripping devices of an industrial robot. It is determined that the horizontal air flow to the capture chamber, compared with that of the vertical, should be used to ensure the maximum possible lifting force. Structures and constriction in fittings, which are used to supply compressed air to the chamber of Bernoulli gripping devices, are considered. Then, the dependencies of the influence of narrowing the fittings on the power characteristics of gripping devices are presented.

show abstract

“…Often the object grasping problem has been focused just on the grasping phase; however, in industrial settings, full pick and place cycle is required. Post gripping [60] is as important as the gripping phase and eventually grasping strategies should be addressed accordingly. For model-free methods, the post-gripping process is extremely complicated and mostly cannot be executed without additional operations after picking up the object.…”

mentioning

confidence: 99%

Smart Industrial Robot Control Trends, Challenges and Opportunities within Manufacturing

2022

View full text Add to dashboard Cite

Industrial robots and associated control methods are continuously developing. With the recent progress in the field of artificial intelligence, new perspectives in industrial robot control strategies have emerged, and prospects towards cognitive robots have arisen. AI-based robotic systems are strongly becoming one of the main areas of focus, as flexibility and deep understanding of complex manufacturing processes are becoming the key advantage to raise competitiveness. This review first expresses the significance of smart industrial robot control in manufacturing towards future factories by listing the needs, requirements and introducing the envisioned concept of smart industrial robots. Secondly, the current trends that are based on different learning strategies and methods are explored. Current computer-vision, deep reinforcement learning and imitation learning based robot control approaches and possible applications in manufacturing are investigated. Gaps, challenges, limitations and open issues are identified along the way.

show abstract

Design of a Deep Post Gripping Perception Framework for Industrial Robots

Cited by 3 publications

References 48 publications

Synthetic Data of Randomly Piled, Similar Objects for Deep Learning-Based Object Detection

Synthetic Data of Randomly Piled, Similar Objects for Deep Learning-Based Object Detection

Influence of Inlet Parameters on Power Characteristics of Bernoulli Gripping Devices for Industrial Robots

Smart Industrial Robot Control Trends, Challenges and Opportunities within Manufacturing

Contact Info

Product

Resources

About