There are some tasks that the human body is just not designed to perform. Unloading trucks and shipping containers is a repetitive, strenuous task – and an enormous reason why the injury rate in warehouses is greater than twice the national average.
The Pickle Robot Company wants its machines to do the heavy lifting. The company's one-armed robots autonomously unload trailers, pick up boxes weighing as much as 50 kilos and place them on on-board conveyors for warehouses of all sorts.
The company name, an homage to The Apple Computer Company, points to the ambitions of founders AJ Meyer '09, Ariana Eisenstein '15, SM '16 and Dan Paluska '97, SM '00. The founders need to make the corporate a technology leader for supply chain automation.
The company's unloading robots mix generative AI and machine learning algorithms with sensors, cameras and vision software to navigate recent environments from day one and improve performance over time. Much of the corporate's hardware is customized by industrial partners. You may recognize the arm from automotive assembly lines, for instance – although chances are you’ll not have seen it in vivid pickle green.
The company is already working with customers similar to UPS, Ryobi Tools and Yusen Logistics to unlock warehouse employees and enable them to unravel other supply chain bottlenecks.
“Humans are really good problem solvers for edge cases, but robots usually are not,” says Paluska. “How can the robot that's really good at brute force and repetitive tasks interact with humans to unravel more problems? The human body and mind are so adaptable, the way in which we perceive and reply to the environment is so adaptable, and robots aren't going to interchange that anytime soon. But there's a lot drudgery we will eliminate.”
Finding problems for robots
Meyer and Eisenstein studied computer science and electrical engineering at MIT, but didn't work together until after graduation, when Meyer founded the technology consulting firm Leaf Labs, which makes a speciality of constructing embedded computer systems for things like robots, cars and satellites.
“This store was run by some friends from MIT,” Meyer recalls, noting that it’s still running today. “Ari worked there, Dan consulted there, and we worked on some big projects. We were the first software and digital design team behind Project Ara, a smartphone for Google, and we worked on a variety of interesting government projects. It was really a life-style company for MIT kids. But after ten years, we thought, 'We didn't come here to do consulting. We got here here to make robots.'”
When Meyer graduated in 2009, problems like robotic dexterity seemed insurmountable. By 2018, the rise of algorithmic approaches similar to neural networks had brought enormous advances in robot manipulation and navigation.
To discover what problem robots could solve, the founders spoke to people from industries as diverse as agriculture, food preparation and hospitality. At some point, they began visiting logistics warehouses and bringing a stopwatch to see how long it took employees to finish various tasks.
“In 2018, we went right into a UPS warehouse and watched 15 people unload trucks during a winter night shift,” Meyer remembers. “We talked to everyone and nobody had worked there longer than 90 days. We said, 'Why not?' They laughed at us. They said, 'Have you ever tried to do that job?'”
It seems that inventory turnover is one among the industry's biggest problems, limiting productivity as managers always struggle with hiring, onboarding, and training.
The founders raised a round of seed funding and built robots that might sort boxes since it was a better task that allowed them to work with technology like grippers and barcode scanners. Their robots eventually worked, but the corporate didn't grow fast enough to be profitable. Worse, the founders had difficulty raising money.
“We were desperately in need of money,” remembers Meyer. “So we thought, 'Why spend our last dollar on a warm-up task?'”
With money becoming increasingly tight, the founders built a proof-of-concept robot that might reliably unload trucks for about 20 seconds and posted a video of it on YouTube. Hundreds of potential customers got here forward. The interest was enough to get investors back on board and keep the corporate afloat.
The company tested its first unloading system in a 12 months with a customer within the California desert, saving human labor from unloading shipping containers that may reach temperatures of as much as 130 degrees in the summertime. Since then, the corporate has expanded deployment to multiple customers and gained traction with third-party achievement centers across the United States
The company's robotic arm comes from German industrial robotics giant KUKA. The robots are mounted on a custom-made mobile base with integrated computer systems, allowing them to autonomously navigate to docks and adjust their position in trailers while lifting. At the top of every arm is a suction cup that holds packages and transports them to the onboard conveyor belt.
The company's robots can accommodate boxes ranging in size from 5-inch cubes to 24-by-30-inch boxes. Depending on their size and weight, the robots can unload between 400 and 1,500 boxes per hour. The company optimizes pre-trained generative AI models and uses a series of smaller models to make sure the robot runs easily in any environment.
The company can be developing a software platform that could be integrated with third-party hardware, from humanoid robots to autonomous forklifts.
“Our immediate product roadmap is loading and unloading,” says Meyer. “But we're also hoping to attach these third-party platforms. Other firms are also attempting to connect robots. What does it mean that the robot unloading a truck communicates with the robot doing palletizing, or that the forklift communicates with the inventory drone? Can they get the job done faster? I believe it's creating an enormous network where now we have to orchestrate the robots and the automation across your entire supply chain, from the mines to the factories to your front door.”
“Why not us?”
The Pickle Robot Company employs about 130 people at its Charlestown, Massachusetts, office, where a standard — if environmentally friendly — office gives technique to a warehouse where robots load boxes onto conveyors alongside human employees and production lines.
This summer, Pickle will ramp up production of a new edition of its system and plans to start developing a two-armed robot sometime later.
“My manager at Leaf Labs once told me, 'Nobody knows what they do, so why don't we?'” Eisenstein says. “I at all times carry that with me. I've been very fortunate to work with so many talented, experienced people in my profession. They all bring their very own skills and understanding. This is a big opportunity – and it's the one way something as difficult as what we do can work.”
The company sees many more robotic problems for its machines in the longer term.
“We didn't say right firstly, 'Let's load and unload a truck,'” says Meyers. “We asked ourselves, 'What does it take to construct an incredible robotics company?' Unloading trucks is the primary chapter. Now now we have built a platform to construct the following robot to assist with more tasks, starting in logistics but then ultimately in manufacturing, retail and hopefully your entire supply chain.”
