AI Picking

To date, industrial robots equipped with vision systems for bulk picking have been put into practical use because the picking patterns of rigid objects such as engineering parts have been standardized. However, when it comes to picking soft or irregular objects such as cables, there are individual differences, so automation was hard to perform and we had to rely on manual labor.

By utilizing the AI technology “Alliom ” that Yaskawa developed, we can create learning data closer to the real environment on the simulator and pick not only rigid objects but also soft objects with the same hand. Since the AI generation process of (1) generation of learning data, (2) learning, and (3) AI generation can be processed on the simulator, the installation time to actual operation, including AI development, is remarkably shortened, and the accuracy of actual operation can also be improve.

For example, in the case of picking parts in bulk, the target parts are brought into the simulator, and AI creates a work environment that is as close to a real environment as possible, including the friction feeling of the parts as well as the angle of the light source in the virtual environment. By using AI to generate a large amount of part data and different ways of stacking parts virtually, the robot hand learns which path and point it can stably hold, and this process is repeated to improve accuracy.

picking parts in bulk

This eliminates the time and cost associated with learning data generated on actual machines, and enables verification and application within 3-4 hours. Ultimately, objects can be taken out with the same hand, from hard parts such as metal to irregular objects such as cables and so on.

In addition, even in processes such as visual inspection of metal casting, where it takes time to turn into big data because defects are hard to come out, realistic defect data is generated from solely dozens of photos by simulation, and implementable AI is generated. We will expand the scope of application of this technology, including the process of inspecting scratches by linking robots and vision cameras.

simulation

Other solutions

  • Production Production
    Production
  • Quality Quality
    Quality
  • Maintenance Maintenance
    Maintenance

Flexible production

High Variety and Variable Quantity Production

High Variety and Variable Quantity Production

By using digital data to manage automated production lines, setup can be prepared automatically without manual intervention, enabling high variety and variable quantity production from a minimum of one unit.

ProductionProduction
 
Autonomous distributed manufacturing

Autonomous Distributed Manufacturing

Digital data such as the torque value, vibration value, and temperature of the servomotor is absorbed into the controller, and the robot can think for itself how to move.

ProductionProduction
 

Accuracy improvement

Accuracy improvement of defect cause analysis Yaskawa case

Accuracy Improvement of Defect Cause Analysis < Yaskawa Case >

By “visualizing” the operation status of equipment/devices with Yaskawa Cockpit, it is possible to identify the root cause by comparing the normal value and abnormal value of the data in the factor analysis for defects in production.

QualityQuality
 

Quality inspection

Automated product quality assessment with AI

Automated Product Quality Assessment with AI

When the quality inspection process is labor-saving, the use of an image judgement service that utilizes AI technology such as deep learning makes it possible to automatically determine complex No Good patterns with the same level of accuracy as humans.

QualityQuality
 

Failure prediction

Predictive failure diagnosis of equipment

Predictive Failure Diagnosis of Equipment

To reduce downtime to zero by performing planned maintenance in anticipation of equipment failure due to wear, etc., in response to concerns that production may become impossible due to the sudden shutdown or something else.

MaintenanceMaintenance
< 

Recovery support

Investigating the cause of equipment failure

Investigating the Cause of Equipment Failure

By acquiring quality data on when, with which equipment, and how it was processed, it is possible to accurately identify the cause of the problem between which equipment and equipment at the time of failure.

MaintenanceMaintenance
 
Faster recovery simulation

Faster Recovery Simulation

The planning technology that Yaskawa developed automatically generates optimal paths, enabling simulation in a few minutes and dramatically reducing engineering time for recovery from sudden stop.

MaintenanceMaintenance