Salinas, Calif.-based Taylor Farms has developed automated harvesters for cabbage (pictured), celery and romaine and is working on others.
Salinas, Calif.-based Taylor Farms has developed automated harvesters for cabbage (pictured), celery and romaine and is working on others.

Increasing automation in the field remains a priority for Taylor Farms.

The company has already improved its efficiency with automated harvesters for romaine, cabbage and celery, allowing it to re-assign workers on those commodities to work on others.

The in-house engineering group now has its sights set on two of the most popular commodities for Taylor Farms.

Broccoli is a big one on our hit list,” said Chris Rotticci, director of automated harvesting equipment for Taylor Farms. “Iceberg is the biggest.”

Broccoli is expensive to harvest because there are multiple cuts involved, Rotticci said, while iceberg lettuce is a target simply because Taylor Farms harvests a massive volume of it daily to meet demand from foodservice customers.

In the upcoming Salinas season, Taylor Farms plans to test automatic iceberg lettuce harvesters.

Though the company sees potential for robotics in the field, creating a machine that performs the necessary functions and is impervious to dirt, rain and bumps in the ground is no easy task. Designers even have to account for natural light.

“All these robots, they work off some kind of optical collaboration, where the optics see a box and then it sends a signal to the computer and the computer tells the robot where to be at what time and what angle,” Rotticci said.

The variety in plants is another challenge.

“You’re not seeing the same romaine in every single field,” Rotticci said. “You’re peeling down three or four wrapper leaves off one head in one bed, and in the next bed you need to take off seven or eight just because the mildew has gotten deeper in that head.”

Along with developing automated harvesting equipment, Taylor Farms engineers are fine-tuning machines currently in use, like those for cabbage and celery.

“We know we can direct-cut those products,” Rotticci said. “Now we’re working on how clean can we get that product to the quality inspectors upstream. If we can clean that head up more for that inspector, they can look at more heads per minute, and it’ll pass quality inspection and in the end provide a cleaner, quality product to the plant, which aids in the processing and shelf life and all that.”

Reducing mechanical damage to plants being harvested is an ongoing goal.

This development takes time. Even when the engineering group has a design and a working prototype, simply building a duplicate machine can take up to six months, depending on the availability of fabricators.

If design time is involved, a project can require nine months to a year, Rotticci said.

Given the expense of developing the machines — particularly with the use of stainless steel for sanitation purposes — a longer timetable is part of the deal.

“We want to hit the ground running,” Rotticci said, “and not have to go backwards too much on the redesign, refabricate side.

“If we have to go a little bit slower from the beginning to have a more consistent run time, I think it will definitely benefit us in the long run,” he said.

While the process is costly, automating the harvest of a crop can pay for itself in two to three years, depending on the commodity and the volume harvested, Rotticci said. Depending on how the machines are used and maintained, they can be used 10 years.