Dextrous Robotics shuts down trailer unloading business


Listen to this article

Voiced by Amazon Polly

Dextrous Robotics, a Memphis-based startup that developed a chopstick-like robot to unload boxes from trailers, has shut down. Evan Drumwright and Sam Zapolsky founded Dextrous in 2019, and the company dissolved in late December 2023.

Dextrous had raised about $8 million in funding since its inception. However, the company was unable to close another round of funding to start production on the commercial version of its DX-1 robot.

“There are so many things that have to go right for a startup to succeed,” Evan Drumwright told The Robot Report. “We knew that going into it. Not that it makes this any easier.”

“If I had to point to one thing, I would point to the venture markets,” he added. “COVID warped investing into startups. When interest rates are low, people have to look for longer-term investments that can pay them better than what they can get through traditional instruments. When interest rates are high, a startup’s short-term profitability has to be higher. It was going to take us a while to build up to profitability.”

Suzanne Drumwright, director of operations at Dextrous Robotics, said the DX-1 robot is completely customized. It used few off-the-shelf parts. This drove the bill of materials (BOM) to more than $300,000 per unit. Evan Drumwright noticed a change in how investors reacted to Dextrous’ BOM in late 2022.

“I don’t remember how many meetings I took where people asked what the BOM was,” said Evan Drumwright, who was a senior research scientist at Toyota Research Institute before founding Dextrous. “Once I said the number, it didn’t matter how that tied into revenue. You could tell they were no longer interested. From 2019 to 2022, we didn’t hear any concern about the BOM whatsoever. But in 2023, that was a different story.

“Investors wanted to know how rapidly our startup could be profitable,” he said. “We stuck with our aggressive, longer-term trajectory, and that didn’t pan out. More than any other factor by far, interest rates being high has changed the investment calculus.”

Hindsight being 20/20, Suzanne Drumwright said Dextrous should have raised more money during the height of the COVID-19 pandemic.

“If we raised more money during COVID, that would’ve been helpful,” she said. “That time was a shining light on the need for this type of application. We were grateful to get seed funding, but we didn’t realize how rare an opportunity it is for a market to spotlight the issues your technology is solving. Now, VCs aren’t writing as big of checks or as many checks.”

The Robot Report reached out to some of Dextrous’ investors for comment but hadn’t heard back by press time.

Dextrous also won a $999,746 grant from the National Science Foundation in September 2023 for a project entitled “Automated Perception for Robotic Chopsticks Manipulating Small and Large Objects in Constrained Spaces.” However, both Evan and Suzanne Drumwright said the grant is paid out in $250,000 chunks over two years and has to be earmarked for certain initiatives.

“You can’t operate on that,” they said.

Evan Drumwright said Dextrous explored being acquired to give the investors some returns. But they didn’t start that process early enough.

“Having never done this before, we didn’t realize people start becoming liable if we continued into insolvency,” he noted. “Before we made people liable, we shut down operations.”

How Dextrous’ robot worked

We first talked to Drumwright in 2020 for an episode of The Robot Report Podcast. At the time, Dextrous exited stealth mode and was ready to discuss its proof of concept, which used Franka Panda cobot arms.

Just six months ago, the company released a video of its second DX-1 prototype. You can watch that video at the top of this page.

The DX-1 features two rigid chopstick-like robot arms that each have five degrees of freedom. The towers that support each arm ride on separate gantries that allow Cartesian movement. The system grabs boxes from multiple sides and applies regulated pressure to grasp packages weighing up to 100 lb., Dextrous said.

Custom sensors enable the DX-1 to detect and move packages. Dextrous said its robot could pick 2,000 parcels per hour, as well as manipulate items as small as a sugar cube and as large as an armchair. Evan Drumwright previously said he envisioned the robots expanding into markets beyond logistics, including agriculture and manufacturing.

Increased competition also a challenge

Another challenge that Dextrous ran into was increased competition. Boston Dynamics, Dexterity, Pickle Robot Co., Rightbot, Mujin, and others are developing robots for trailer loading and unloading. Evan Drumwright said several investors who declined to provide funding told Dextrous they were going to wait and see how the competition unfolds.

Boston Dynamics has found early success with its Stretch robot that is being deployed by the likes of DHL, NFI, Otto Group — which were all multi-million dollar contracts.

Pickle recently announced Yusen Logistics as a customer. Yusen said it unloads 200,000 containers annually. Mujin has raised $104 million in Series C funding, and Rightbot picked up $6.25 million from Amazon.

Those aforementioned systems all use a variation of suction-based grippers to manipulate boxes, unlike Dextrous’s chopstick-like manipulators that grasped each box from two sides. Each approach has its pros and cons, but perhaps the chopstick approach was too different.

“The chopstick approach was fairly obvious,” recalled Evan Drumwright about how Dextrous landed on this approach. “If you talk about humanoids, their hands are just not sufficient. Those anthropomorphic hands are far less capable than a human hand. So you need another type of tool.”

“And the pinching robot grippers won’t work here either because of the size and shape disparity of the boxes,” he said. “So this is why people went with suction, which has its own pros and cons. Intelligence and chopsticks can lift a lot of things — our thesis was proven.”

There is demand for these types of robots. Unfortunately, Dextrous couldn’t grasp the opportunity.

Strategically located in Memphis

Dextrous strategically picked Memphis for its headquarters. Memphis is one of the leading logistics hubs in the U.S. Dextrous hoped its robots would be used by the national and international logistics companies based there. In May 2023, Dextrous relocated to a larger building in Midtown Memphis.

Obviously, things didn’t go according to plan.

Evan Drumwright claimed Dextrous had a healthy company pipeline within logistics and retail.

“Nobody has a solution to this entire problem,” he said. “That’s not to say we’d be a perfect fit for everybody. For some customers, the suction approach is perfect. But the majority of retailers, manufacturers, 3PLs, and package logistics companies deal with some much variation in size, shape, and payload that there wasn’t a solution in existence or on the horizon that could compete with us.”

“We’re both bummed, but the overwhelming feeling after sitting on it for a few weeks is being proud,” said Suzanne Drumwright. “With under $10 million, a team of 10 built a robot that uses chopsticks to unload shipping containers. Is it too innovative, is it too different? We heard lots of cheering from the sidelines that it worked.”

Evan Drumwright said they’ve secured Dextrous’ IP back from investors. As for what’s next? They’re not exactly sure, but another robotics startup is certainly not out of the question.

“We’re super proud of what we built,” Evan Drumwright said. “Going into it, we wanted to build something using this chopstick approach and move things at high speeds in dynamic environments. And we did it. I would love it if we were focused on fulfilling contracts with the big customers we had lined up, but the fact we put this together and made it work like we hoped to is a good second prize.”

Latest articles

spot_imgspot_img

Related articles

Leave a reply

Please enter your comment!
Please enter your name here

spot_imgspot_img