RGB-D Tracking and Optimal Perception of Deformable Objects

Cuiral-Zueco, Ignacio; López‐Nicolás, Gonzalo

doi:10.1109/access.2020.3012067

Cited by 6 publications

(5 citation statements)

References 33 publications

(39 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Once the outsole has been placed on the shoe, the second type of movement begins. To perform this task, target points over the shoe upper for each of the grippers must be calculated, and the movement of both robots has to be controlled [38] with the help of the 3D vision system [39], [40] that has been installed in the structure of the system. This movement, despite requiring high precision in the target location, is a slow and short-distance movement, therefore, synchronization is not extremely relevant.…”

Section: ) Assembling Stationmentioning

confidence: 99%

See 1 more Smart Citation

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

Section: ) Assembling Stationmentioning

confidence: 99%

“…To ensure that the outsole is correctly joined to the upper the system tracks [39] some points of the outsole using the 3D vision system to adapt the shape to the upper part of the shoe. When assembly is finished, and with the aim of strengthening the joint, a final pressing step is performed, using the robots' own grippers.…”

Section: ) Assembling Stationmentioning

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

show abstract

“…For deformable object perception, previous works learn and leverage appearance and geometric information [36,33,40], softness [3], visual-tactile sensation [7], shape completion [4], emotional perception [24], optimal perception [8] and the continuity of perception [29,9,22]. The manipulation is challenging for its high complexity.…”

Section: Perceive and Manipulate Deformable Objectsmentioning

confidence: 99%

Learning Foresightful Dense Visual Affordance for Deformable Object Manipulation

Wu¹,

Ning²,

Dong³

2023

Preprint

View full text Add to dashboard Cite

Understanding and manipulating deformable objects (e.g., ropes and fabrics) is an essential yet challenging task with broad applications. Difficulties come from complex states and dynamics, diverse configurations and highdimensional action space of deformable objects. Besides, the manipulation tasks usually require multiple steps to accomplish, and greedy policies may easily lead to local optimal states. Existing studies usually tackle this problem using reinforcement learning or imitating expert demonstrations, with limitations in modeling complex states or requiring hand-crafted expert policies. In this paper, we study deformable object manipulation using dense visual affordance, with generalization towards diverse states, and propose a novel kind of foresightful dense affordance, which avoids local optima by estimating states' values for longterm manipulation. We propose a framework for learning this representation, with novel designs such as multi-stage stable learning and efficient self-supervised data collection without experts. Experiments demonstrate the superiority of our proposed foresightful dense affordance. Project page: https://hyperplane-lab.github.io/DeformableAffordance * Equal contribution, order determined by coin flip.

show abstract

“…Herguedas et al examined the optimal sensor placement for deformable bodies [ 50 ]. The procedure was later improved using RGB-D cameras [ 51 ]. An important area of the Optimal Sensor Placement Problem is the vibration measurements in various structures such as bridges [ 52 ].…”

Section: Related Literaturementioning

confidence: 99%

Placement of Optical Sensors in 3D Terrain Using a Bacterial Evolutionary Algorithm

Kovács

Bolemányi

Botzheim

2022

Sensors

View full text Add to dashboard Cite

This paper proposes an optimization framework for terrain large scale optical sensor placement to improve border protection. Compared to the often used, maximal coverage of an area approach, this method minimizes the undetected passages in the monitored area. Border protection is one of the most critical areas for sensor placement. Unlike traditional border protection solutions, we do not optimize for 2D but for 3D to prevent transit. Additionally, we consider both natural and built environmental coverings. The applied environmental model creates a highly inhomogeneous sensing area for sensors instead of the previously used homogeneous one. The detection of each sensor was provided by a line-of-sight model supplemented with inhomogeneous probabilities. The optimization was performed using a bacterial evolutionary algorithm. In addition to maximizing detection, minimizing the number of the applied sensors played a crucial role in design. These two cost components are built on each other hierarchically. The developed simulation framework based on ray tracing provided an excellent opportunity to optimize large areas. The presented simulation results prove the efficiency of this method. The results were evaluated by testing on a large number of intruders. Using sensors with different quantities and layouts in the tested 1×1×1 km environment, we reduced the probability of undetected intrusion to below 0.1% and increased the probability of acceptable classification to 99%.

show abstract

RGB-D Tracking and Optimal Perception of Deformable Objects

Cited by 6 publications

References 33 publications

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

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

Learning Foresightful Dense Visual Affordance for Deformable Object Manipulation

Placement of Optical Sensors in 3D Terrain Using a Bacterial Evolutionary Algorithm

Contact Info

Product

Resources

About