Integration of a reconfigurable robotic workcell for assembly operations in automotive industry

Nemec, Bojan; Mavsar, Matija; Simonič, Mihael; Hrovat, Matevž Majcen; Škrabar, Jure; Ude, Aleš

doi:10.1109/sii52469.2022.9708896

Cited by 5 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When we displaced the target position by 5 mm, the robot correctly classified the context to be “x/-”, whereas the magnitude was not determined precisely (4 mm) as we used the model that did not include error contexts of this magnitude in the training set. Nevertheless, by combining the policies demonstrated for the other two cases in the “x/-” category and using locally weighted regression, as proposed in [ 6 , 43 ], the robot was able to perform the insertion successfully.…”

Section: Resultsmentioning

confidence: 99%

“…In our research, we focused on robotic assembly and considered two tasks—square peg insertion using the Cranfield benchmark [ 42 ] and the industrially relevant car starter assembly [ 43 ]. To perform both tasks and collect data for the context determination model, we rely on a modular workcell design that enables easy mounting of task-specific equipment, e.g., robots, sensors, and auxiliary devices [ 44 ] and a ROS-based software architecture that allows for easy integration of new components [ 45 ].…”

Section: Methodsmentioning

confidence: 99%

“…Previous automation attempts have failed due to the low success rate that was achieved. To ensure robust insertion, we proposed to use the exception strategy framework [ 43 ].…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Determining Exception Context in Assembly Operations from Multimodal Data

Simonič

Hrovat

Džeroski

et al. 2022

Sensors

View full text Add to dashboard Cite

Robot assembly tasks can fail due to unpredictable errors and can only continue with the manual intervention of a human operator. Recently, we proposed an exception strategy learning framework based on statistical learning and context determination, which can successfully resolve such situations. This paper deals with context determination from multimodal data, which is the key component of our framework. We propose a novel approach to generate unified low-dimensional context descriptions based on image and force-torque data. For this purpose, we combine a state-of-the-art neural network model for image segmentation and contact point estimation using force-torque measurements. An ensemble of decision trees is used to combine features from the two modalities. To validate the proposed approach, we have collected datasets of deliberately induced insertion failures both for the classic peg-in-hole insertion task and for an industrially relevant task of car starter assembly. We demonstrate that the proposed approach generates reliable low-dimensional descriptors, suitable as queries necessary in statistical learning.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Determining Exception Context in Assembly Operations from Multimodal Data

Simonič

Hrovat

Džeroski

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…However, these systems often lack the precision required for the proposed operation. In the automotive industry, numerous assembly operations can be automated using robots [13]. Nevertheless, the majority of components manipulated by these robots are rigid and lack flexibility.…”

Section: Introductionmentioning

confidence: 99%

Development of an Integrated Robotic Workcell for Automated Bonding in Footwear Manufacturing

Gómez-Hernández,

Gutiérrez-Hernández,

Jimeno-Morenilla

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Traditional manufacturing industries are currently immersed in an automation process, integrating new techniques and tools, driven by the demands from producers to improve the manufacturing process as well as the working conditions of employees.For the footwear industry, bonding is a key operation in the manufacturing process where the outsole is assembled onto the lasted shoe. However, in this operation, workers are often subjected to hazardous substances (i.e., organic solvents) and perform repetitive tasks with limited added value.Against this background, this paper describes the results of a research project, whose aim was to obtain the maximum benefit from different technologies analyzed, such as collaborative robotics, artificial vision and multirobot control, for the manipulation of flexible/deformable objects.The main result of this project is a robotic workcell for shoe bonding that has been introduced in the production line to fully automate the operation. This workcell integrates three collaborative robots, one for (hot melt) adhesive application and another two, with two-finger electric grippers, to carry out the bonding synchronously. Different vision systems have also been embedded to conduct the various processes involved. The entire operation is controlled and coordinated through ROS (Robot Operating System).The key findings of this research showcase the automation of a process traditionally undertaken by humans. In this novel approach, two robots collaborate to manipulate flexible objects, liberating the operator from engaging in repetitive, non-value-added tasks and the handling of hazardous substances.

show abstract

“…In this way, the robotic manipulator is provided with the information required to ensure human safety and maintain efficiency. The activity recognition module provides the contextual information required for pose extrapolation, thus decreasing the system’s reaction time, similar to the work described in [ 43 , 44 ]. The difference between these two approaches is the information input to the AR module and the selection of the output classes.…”

Section: Introductionmentioning

confidence: 99%

Towards a Safe Human–Robot Collaboration Using Information on Human Worker Activity

Orsag

Stipančić

Koren

2023

Sensors

View full text Add to dashboard Cite

Most industrial workplaces involving robots and other apparatus operate behind the fences to remove defects, hazards, or casualties. Recent advancements in machine learning can enable robots to co-operate with human co-workers while retaining safety, flexibility, and robustness. This article focuses on the computation model, which provides a collaborative environment through intuitive and adaptive human–robot interaction (HRI). In essence, one layer of the model can be expressed as a set of useful information utilized by an intelligent agent. Within this construction, a vision-sensing modality can be broken down into multiple layers. The authors propose a human-skeleton-based trainable model for the recognition of spatiotemporal human worker activity using LSTM networks, which can achieve a training accuracy of 91.365%, based on the InHARD dataset. Together with the training results, results related to aspects of the simulation environment and future improvements of the system are discussed. By combining human worker upper body positions with actions, the perceptual potential of the system is increased, and human–robot collaboration becomes context-aware. Based on the acquired information, the intelligent agent gains the ability to adapt its behavior according to its dynamic and stochastic surroundings.

show abstract

Integration of a reconfigurable robotic workcell for assembly operations in automotive industry

Cited by 5 publications

References 23 publications

Determining Exception Context in Assembly Operations from Multimodal Data

Determining Exception Context in Assembly Operations from Multimodal Data

Development of an Integrated Robotic Workcell for Automated Bonding in Footwear Manufacturing

Towards a Safe Human–Robot Collaboration Using Information on Human Worker Activity

Contact Info

Product

Resources

About