Photo credit: www.therobotreport.com
Automotive manufacturing has seen substantial advancements through the strategic application of robotics and automation technologies. These innovations are particularly relevant in executing complex tasks that are often burdensome or unsafe for human employees. One such instance is the recent collaboration between DENSO, a prominent global automotive parts manufacturer, and CapSen Robotics, aimed at automating a demanding tote-handling process.
In a bid to enhance productivity and workplace safety, DENSO sought to alleviate the physical strain on its workforce, which was engaged in lifting and transporting hefty stacks of totes to and from a paint booth. To address this challenge, DENSO turned to CapSen Robotics, which tailored its sophisticated 3D vision, motion planning, and control software to meet the specific requirements of the task.
Additionally, DENSO collaborated with Invent Automation to design a system aimed at optimizing workflow, thus enabling employees to focus on more critical and less physically taxing roles. This initiative is part of DENSO’s overarching strategy to augment human labor with automation, enhancing both employee satisfaction and overall operational efficiency.
Streamlining Operations at DENSO
As part of its production process, DENSO applies a low-gloss charcoal paint to automotive parts, giving them an aesthetically pleasing finish while minimizing glare within vehicle interiors.
Previously, the procedure involved workers manually stacking six heavy totes filled with plastic automotive components onto a conveyor. An operator would then remove the components, transport them to the painting station, and repack them into the totes — a process that required an employee to cycle through stacking and unstacking every 30 seconds. Recognizing the inefficiencies of this repetitive task, the company sought an automation solution to optimize productivity and reduce physical strain on employees.
In response to space limitations on the factory floor that hindered conventional automation methods, a compact and flexible robotic system was developed. This innovative setup features an inbound conveyor that transports the totes to a six-axis collaborative robot equipped with an Intel RealSense 3D RGB depth camera. This camera is integral to the robot’s ability to visualize and evaluate the tote’s dimensions for effective handling.
The CapSen PiC 2.0 software, which operates on an industrial PC with GPU capabilities, allows the robot to plan its movements accurately, pick up, and transition the totes to the painting area. Here, the automotive parts undergo a painting process, followed by curing and inspection before being returned to the totes for conveyance back to the robot.
Addressing Challenges in Automation
One of the project’s primary concerns was ensuring that the robot could handle the totes accurately without risking collisions during its operation. To fine-tune the robot’s movement capabilities, CapSen Robotics employed the PiC 2.0 simulation environment, which proved vital for assessing the system’s reachability and collision avoidance capabilities.
Despite these precautions, real-world tests revealed a new complication: the totes began to bend when lifted. To counter this issue, the project team worked collaboratively to refine the system design. CapSen Robotics utilized artificial intelligence in their software to enhance the robot’s performance in diverse pick-and-place tasks, including this tote-handling application. This necessitated customizing their AI models to adapt to the specific challenges posed by the production environment.
In particular, CapSen’s AI was programmed to recognize and accurately locate the tote handles through machine learning methodologies, thus improving pick reliability. Instead of relying on traditional scales to gauge the tote’s weight, the robot’s torque sensors provided real-time weight assessments, minimizing the risk of abrupt movements that could lead to dropped totes.
Additionally, the design team faced challenges concerning the stacking of totes. With up to six stacked totes, improper alignment could result in operational delays. To prevent this, the system includes a feature where the robot slightly retreats after placing a tote to visually confirm its proper positioning through the camera, ensuring that the stack is secure before proceeding.
DENSO’s Automation Success Story
The adaptability of CapSen PiC 2.0 was pivotal in this endeavor, providing a robust platform that harnesses data from various hardware components without the limitations of traditional programmable logic controllers (PLCs). The capabilities offered by the CapSen human-machine interface (HMI) further streamlined operations, allowing non-technical staff to manage the system efficiently while visualizing real-time data, including camera feeds.
CapSen’s software, being hardware agnostic, afforded DENSO the flexibility to incorporate automated mobile robots (AMRs) during testing to enhance logistics within the facility. This integration not only simplified tote movement but also exemplified the system’s scalability for future applications.
Since the implementation of this advanced automation system, DENSO has reported outstanding results, achieving a flawless record with no dropped totes or missed picks. This success has significantly alleviated the physical demands on employees, marking a successful transition to a more efficient and safer workplace.
Source
www.therobotreport.com