Automotive Industry Eyes 3D Vision Guidance with Robots

Jordan Merhib, vice president of ISRA Vision Systems (Lansing, MI) says that through the use of vision guidance with robots, companies are able to solve tough problems without breaking the budget. Mistakes are costly, Merhib notes, in terms of both dollars and reputation, and 3D machine vision guidance with robots provides quality control that is impossible to match using the human eye or hand.

“The use of machine vision with robots to assemble boards in the semiconductor industry drove quality to higher levels. Before the use of machine vision, there was a 10 percent fallout,” said Merhib. “The companies using the boards asked, ‘What’s this garbage they’re sending us?’ The same dynamic is driving the use of vision guidance with robots in the automotive industry. It is improving the quality of the product.”

As the automotive manufacturers re-think their lines and processes, says Merhib, he sees two recurring objectives: reduce costs and increase flexibility.

“3D vision provides a means to accomplish both of these seemingly contradictory goals,” he says. “A hard-tooled station can typically accommodate only one type of product. Visual fixturing, on the other hand, allows you to process multiple products down the same line so the flexibility goes up and the cost of re-tooling for new products goes down.”

Additionally, says Merhib, “the cost of maintaining a vision system over time is practically nil while a hard-tooled fixturing solution requires increasing maintenance due to wear.”

Auto Racking and Bin Picking
Common applications of 3D vision robotic guidance within the automotive industry are auto racking – the use of robots to pick parts out of racks – and bin picking – using robots to pull parts out of bins, says Adil Shafi, president of Shafi Inc., (Brighton, MI).

According to Shafi, one of the most significant benefits of automated racking is when stamping and assembly-by-welding, which can be quick operations, but are delayed by the manual loading or unloading of parts, such as after stamping, before welding assembly or after welding assembly. Robots, using vision guidance, are able to pick out the best parts for each application, and place them exactly where they need to be for optimum assembly.

“Normally these solutions are justified [with the savings to] labor costs,” says Shafi. But, he adds, there are additional benefits, such as faster production times because manual loading and unloading of parts is eliminated, reduced safety problems because employees are no longer lifting and carrying parts, and not having to retool racks when product geometries change for new product launches. The robots can be reprogrammed at a fraction of the time and cost it takes to build a new production line.

Robotic Guidance and Visual Inspection
Mike Blood, product marketing manager at Perceptron, (Plymouth, MI) said his customers are turning to 3D vision guidance because it allows them to track product quality in real time. “Whenever a framed car body goes through the vision station anywhere from 30 to 120 sensors measure key characteristics critical to design and build,” he says. “You can sample one or two jobs a day or you can measure every vehicle produced. The customer achieves better process control by immediately analyzing data from the last 10 jobs or the last 1,000 jobs. You can make proactive moves right away, rather than be reactive a week from now when you’re trying to discover what went wrong on the line.”

Robot guidance systems, he said, add to quality control because the same 3D vision sensors mounted on the robots are able to measure the roof opening on the car body as well as the roof.

“The robots have four sensors mounted on an end effector, and the roof is measured before the robot even picks it up. Then the robot holds the roof over the vehicle – about 50 mm above – and the same four sensors measure the roof opening. Calculations and transformations are made and the robot is told how to correctly adjust the position of this particular roof for this particular opening to achieve optimal margins and flushness,” says Blood.

A second use for 3D vision guidance with robots, says Blood, includes utilizing robot-mounted sensors on either side of an assembly line for bare metal or painted cars to measure closure margin and flush. And in some cases, direct final line fitters. The robots, he notes, can be adjusted for line speed and the color of the car. A third use, Blood adds, is to scan features like the outer edge of a car door to check fit. Before 3D machine vision, all of these checks were done manually.

“With vision guidance and robots, you achieve accuracy and repeatability,” says Blood. “Competitively, this is an advantage. Quality is all about being able to repeat results.”

 The author invites questions about this article.