How Locus is getting a grasp on one of robotics biggest challenges: manipulation

NeuraGrasp can grasp a variety of things, from small containers to porous polybags and material objects. | Supply: Locus Robotics

Locus Robotics Corp. has efficiently tackled many robotic challenges in actual working environments. Its autonomous cell robots, or AMRs, have navigated numerous miles in quite a lot of services, and so they have aided in additional than 6 billion picks. Now, its concentrating on manipulation.

The corporate entered the robotic selecting area in March 2025 when it unveiled Array, a cell manipulator outfitted with a custom-developed, vision-guided selecting arm. Over a yr later, Locus acquired Nexera Robotics, which is understood for its NeuraGrasp gripping know-how.

In response to Roy Belak, the senior vice chairman of robotic greedy at Locus and former CEO of Nexera, by the point it got here involved with Locus, Nexera had already spent years refining its distinctive, smooth finish effectors. It had already gone by way of at the least six completely different product generations, he mentioned.

“Simply as we had been attending to that time the place the know-how was displaying its compelling capabilities, we had been at a fortuitous scenario the place we met with Locus at MODEX, and I believe the historical past type of writes itself from there,” Belak instructed The Robotic Report.

The timing was very best for each firms. Locus needed to have the ability to deal with extra SKUs in its services, and Vancouver, Canada-based Nexera was able to scale.

“We need to see this know-how deployed at scale,” mentioned Belak. “That’s one thing that excites me as anyone who’s an innovator and constructing one thing. You simply need to see it on the market extensively, and that appeared prefer it was a terrific alternative.”

What does NeuraGrasp carry to Locus Array?

Nexera met Locus in 2025, when Locus was planning its early market launch for Array. On the time, Array had a regular suction-cup gripper. Belak mentioned the businesses spent plenty of time determining how you can match Nexera’s know-how into Array’s growth so late within the sport.

Fortunately, though Locus was already far into its technique of creating Array, the robotic was already nicely ready for Nexera’s know-how. Array already had an air provide, an influence provide, and different issues that Nexera’s know-how wanted.

“There was plenty of work that was finished previous to the acquisition to verify the know-how may very well be slipped into the event stream,” recalled Belak. “I believe it we had missed that window by one other yr, it most likely would have been actually arduous to make these adjustments.”

Whereas Belak didn’t give an actual timeline of when clients may acquire entry to Nexera’s know-how, he mentioned he’s optimistic concerning the course of. Belak mentioned he expects that the subsequent time Array is displayed publicly, it should doubtless have Nexera know-how.

Proper now, Locus is within the early levels of deploying Array with its authentic gripper to clients, so Nexera hasn’t seen the windfall of knowledge that will likely be coming quickly.

“There’s an extended line of shoppers with their fingers up which might be actually enthusiastic about Array,” Belak mentioned. It’s in some methods intimidating, since you’re going to take this factor, you’re going to combine it, after which it’s  going to go to many various locations. Then we’ll begin seeing the info actually come again, and we’ll have an opportunity to simply optimize as a lot as we are able to throughout that course of.”

Suction versus pinch selecting

When taking a step again and looking out on the robotics business as a complete, Belak mentioned that a lot of robotic manipulation falls on a spectrum between suction-based greedy and pinch greedy.

“The fact is that suction is superb, significantly in warehouse operations,” he mentioned. “Sixty p.c to 70% of the time, it’s all you want. It’s whenever you get to that 30% to 40% and get to the actually excessive protection ratios.”

NeuraGrasp’s smooth gripping know-how can assist to bridge this hole as a result of it makes use of each suction and pinching, in line with Belak.

“As an business development, it’s 100% acknowledged that pinch greedy is probably going going to be very influential in the way forward for robotic greedy,” he mentioned. “With that mentioned, we’re nonetheless agency believers that suction will get you portion of the best way with air. And if it ain’t broke, you don’t throw it away.”

Locus is specializing in reliability and high quality picks, not velocity

“Within the industrial surroundings, reliability is all the pieces,” Belak mentioned. “Suction applied sciences, though they’ve a efficiency limitation, the fantastic thing about them is that they’re extremely dependable. As we go to pinch greedy applied sciences, relying on the extent of complexity that’s there, they’ll border on marginally dependable to very unreliable, or perhaps simply questionably dependable.”

Belak mentioned it’s necessary to make robots that may merely do duties efficiently. However after that, the highest precedence is reliability, as a result of it doesn’t matter how nicely a robotic can choose if it’s always breaking down or lacking picks. As well as, the standard of the choose is important, he mentioned.

“There’s an enormous differentiation between having the ability to do one thing, and having the ability to do the identical duties with top quality. Locus Origin bots are already recognized for being very high-quality,” Belak mentioned. “We need to carry that very same ethos to selecting.”

“Choosing has traditionally been a discipline the place the claims have been extra targeted on the return on funding, with out understanding the underlying high quality,” he continued. “We will do many picks per hour, however what folks don’t concentrate on is how good are these picks.”

Locus’ staff needs to concentrate on avoiding product harm, double picks, and misspicks, famous Belak. They aren’t typically talked about when promoting a robotic, however they’ll eat right into a buyer’s prices rapidly, he added. “As soon as they’ve the standard down, then they’ll fear about rising velocity,” Balak mentioned.

What’s holding again robotic selecting?

Nexera’s NeuraGrip can grasp items that were previously difficult to pick, such as porous polybags.

Nexera’s NeuraGrip can grasp objects that had been beforehand troublesome to select, akin to porous polybags. | Supply: Locus Robotics

Two roadblocks are nonetheless hindering the business, claimed Belak: the event of AI and giving robots a way of contact. The business is sweet at constructing fancy fingers or electromechanical buildings, he mentioned. However, in terms of pinch greedy, the intelligence remains to be lagging behind.

“[The challenge] is 100% the proprioception and the info metrics and the suggestions mechanisms which might be there to truly make these fingers do issues,” Belak mentioned.

This shouldn’t come as a shock. The robotics business has raked in tens of millions of {dollars} of funding in recent times to construct a strong bodily AI system for robots. And the truth is, human brains developed over billions of years. Some firms try to shut this hole by producing as a lot information as attainable, however whenever you do this, you begin to discover the dearth of tactile sensing.

“There’s one piece of the bodily system which is lacking, which is the sense of contact on the robots, and significantly sturdy, dependable contact that doesn’t degrade after very restricted utilization. That stuff is actually powerful to simulate or to have the info generated,” Belak mentioned. “I don’t suppose that simulated information goes to assist with this drawback very a lot.”

Nexera does at present use a way of contact, and it’s fascinated about advancing that sooner or later. Whereas Locus remains to be early in its journey with Array, Belak mentioned the corporate’s manipulation roadmap doesn’t cease with Array.

“Robotic manipulation typically — whether or not it’s Array, whether or not it’s different variations of the know-how, whether or not it’s on selecting benches, no matter it could be — we’ve got recognized that as a pillar of innovation that we need to be distinctive at,” he mentioned.

Editor’s be aware: Rick Faulk, the CEO of Locus Robotics, will likely be talking at RoboBusiness 2026 in Santa Clara, Calif., on Oct. 20 and 21. Faulk will likely be a part of the panel dialogue “The RaaS Playbook: Pricing, Service & Scale.”



SITE AD for the 2026 RoboBusiness call for speakers
Register now and save in your go to RoboBusiness 2026

The submit How Locus is getting a grasp on one among robotics largest challenges: manipulation appeared first on The Robotic Report.