Tech
Robots trained with spatial dataset show improved object handling and awareness
When it comes to navigating their surroundings, machines have a natural disadvantage compared to humans. To help hone the visual perception abilities they need to understand the world, researchers have developed a novel training dataset for improving spatial awareness in robots.
In new research, experiments showed that robots trained with this dataset, called RoboSpatial, outperformed those trained with baseline models at the same robotic task, demonstrating a complex understanding of both spatial relationships and physical object manipulation.
For humans, visual perception shapes how we interact with the environment, from recognizing different people to maintaining an awareness of our body’s movements and position. Despite previous attempts to imbue robots with these skills, efforts have fallen short as most are trained on data that lacks sophisticated spatial understanding.
Because deep spatial comprehension is necessary for intuitive interactions, if left unaddressed, these spatial reasoning challenges could hinder future AI systems’ ability to comprehend complex instructions and operate in dynamic environments, said Luke Song, lead author of the study and a current Ph.D. student in engineering at The Ohio State University.
“To have true general-purpose foundation models, a robot needs to understand the 3D world around it,” he said. “So spatial understanding is one of the most crucial capabilities for it.”
The study was recently given as an oral presentation at the Conference on Computer Vision and Pattern Recognition. The work is published in the journal 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
To teach robots how to better interpret perspective, RoboSpatial includes more than a million real-world indoor and tabletop images, thousands of detailed 3D scans, and 3 million labels describing rich spatial information relevant to robotics. Using these vast resources, the framework pairs 2D egocentric images with full 3D scans of the same scene so the model learns to pinpoint objects using either flat-image recognition or 3D geometry.
According to the study, it’s a process that closely mimics visual cues in the real world.
For instance, while current training datasets might allow a robot to accurately describe a “bowl on the table,” the model would lack the ability to discern where on the table it actually is, where it should be placed to remain accessible, or how it might fit in with other objects. In contrast, RoboSpatial could rigorously test these spatial reasoning skills in practical robotic tasks, first by demonstrating object rearrangement and then by examining the models’ capacity to generalize to new spatial reasoning scenarios beyond their original training data.
“Not only does this mean improvements on individual actions like picking up and placing things, but also leads to robots interacting more naturally with humans,” said Song.
One of the systems the team tested this framework on was a Kinova Jaco robot, an assistive arm that helps people with disabilities connect with their environment.
During training, it was able to answer simple close-ended spatial questions like “Can the chair be placed in front of the table?” or “Is the mug to the left of the laptop?” correctly.
These promising results reveal that normalizing spatial context by improving robotic perception could lead to safer and more reliable AI systems, said Song.
While there are still many unanswered questions about AI development and training, the work concludes that RoboSpatial has the potential to serve as a foundation for broader applications in robotics, noting that more exciting spatial advancements will likely branch from it.
“I think we will see a lot of big improvements and cool capabilities for robots in the next five to ten years,” said Song.
Co-authors include Yu Su from Ohio State and Valts Blukis, Jonathan Tremblay, Stephen Tyree and Stan Birchfield from NVIDIA.
More information:
Chan Hee Song et al, RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Robotics, 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2025). DOI: 10.1109/cvpr52734.2025.01470
Citation:
Robots trained with spatial dataset show improved object handling and awareness (2025, November 13)
retrieved 13 November 2025
from https://techxplore.com/news/2025-11-robots-spatial-dataset-awareness.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
Tech
The Aventon Soltera 3 Is the Most Bikey Ebike on the Market Right Now
Belt-drive bikes offer some huge upsides. First, they usually require less maintenance, with many belts often lasting twice as long as a typical chain. Second, there’s no grease to speak of, and therefore, no black smudges on your work pants. Third, in the case of the Soltera 3, the belt comes from the Gates brand, whose drivetrain belts are as good as it gets. Belt-drive bikes are silent and often smoother than their chain-driven counterparts.
That said, the inclusion of a low-maintenance element such as a belt drive paired with hydraulic disc brakes, which require bleeding roughly every year, struck me as an odd choice. If Aventon wanted to make the Soltera 3 as hands-off as possible, cable-actuated brakes would have been a more intuitive choice.
The other thing that immediately jumps out about the Soltera 3 is its relatively light weight. At 37 pounds, the Soltera 3 is heavy for an analog bike. But it’s certainly not heavy for an ebike, and it’s nearly as stiff, nimble, and navigable as a conventional bicycle. One issue I’ve always had with ebikes is their heft. Given that they’re often made to replace a car, they’re built with load bearing in mind. Also, ebike batteries are heavy.
Adding to that sense of “this is just like my other bikes,” the Soltera 3 simply looks cool, which is often not the case when it comes to ebikes. The matte black my tester bike arrived in looks cool because matte black almost never doesn’t look cool. (Additionally, the Soltera 3 is available in dark matte blue and a sleek silver.) But beyond the finish, the bike’s geometry; its wide, almost perfectly flat handlebars; and its narrow (by ebike standards) 700 x 36 tires make it feel closer in DNA to a road bike than a traditional ebike.
Button Press
Photograph: Michael Venutolo-Mantovani
I’m 6′4′′, and the extra large Soltera 3 that I tested was at a maximum saddle height. It was suitable for me, but I couldn’t recommend anyone bigger than me riding the Soltera 3. That said, with four sizes ranging from small to extra large, the line covers a wide swath of riders, ranging from my height all the way down to 5′ tall.
Tech
Tin Can Is a Dumb Phone for Kids. Can Someone Teach Them How to Use It?
Chet Kittleson, 38, is the cofounder of Tin Can and a father of three kids, 10, 8, and 5. I suspect he wouldn’t much like my description of the product’s function as “spying” (keeping watch over one’s kids is part of a parent’s job) or the product itself as a “toy.” He thinks of it, instead, as a utility: a way for kids to talk to Grandma or make plans with friends and to be “part of the same world that grown-ups are a part of.” When he was a kid, he says, the landline was “arguably the most successful social network of all time.” Every house had one. Then came cell phones and smartphones. Direct lines to the internet. “And somewhere along the way we decided the landline was obsolete,” Kittleson says. “In doing that, we overlooked a group that was a major beneficiary of it: kids.”
I’m talking to him over Zoom one afternoon from my home in Los Angeles and his office in Seattle. When I tell him that Amos and Clara had called me more than two dozen times, he doesn’t seem particularly surprised. At first there’s a burst of activity, he says, and then over the course of a few weeks, the kids mature. “They’re like, oh, OK, I see that I can actually do things with this that are important,” he says.
Kittleson, who guesses that most Tin Can users are between the ages of 5 and 13, says he wants to help create a “better childhood” or, as he puts it, “giving kids back a sense of independence and confidence.” (Mike Duboe, a partner at Greylock Ventures, which led a round that invested $12 million in the company in October, says something similar.) One parent, describing their kid’s Tin Can use on X, wrote that it “felt like the old days.”
Amos and Clara weren’t the only ones who, over the holidays, got the gift of gab. In late December, frustrated parents flooded the company’s feedback forms and posted on Reddit that their Tin Cans weren’t working. Though the Tin Can engineers had anticipated a surge in usage around the holidays, the hundredfold increase in call volume took them by surprise.
When I ask Kittleson about the holiday meltdown, he winces. “It was a stressful Christmas,” he concedes. (A message on the Tin Can homepage said, “We’re investigating an issue impacting the network.”) He says that future shipments of the product will be staggered.
And the product’s far from perfect: There can be echoes, unstable sound quality, and long pauses. The buttons on the device are hard to press, which can be challenging to little fingers like Amos’. His mother, Rebecca, sometimes has to help him make calls. “It takes a little bit out of the independence of it,” she says.
My first phone, like that of other kids in my generation, was my family’s, a mustard yellow piece of hard plastic that sat on the mottled brown linoleum counter adjacent to the kitchen. It held a special place in my imagination—an object full of potential—but like most phones back then it was shared within a family and maybe even overheard or monitored. It was also tethered to a wall, making it difficult to multitask or move around while on a call. Kittleson, in fact, says that one inspiration for Tin Can was his frustration when he called his mother on her cell phone. She was, he says, “the worst”: the sort of person who ran around the house while on the call, doing laundry or whatnot. Difficult to hear. Easily distracted.
Tech
The Newest Google Pixel Phone Comes With a $100 Gift Card (for Now)
The Google Pixel 10a comes out on March 5, and right now, Amazon and Best Buy are both offering free $100 gift cards when you preorder. The phone’s price is $499, but the gift card sweetens the deal. It’ll be automatically added to your cart.
Amazon’s gift card is a physical one that ships with the phone; Best Buy’s is a digital gift card that will be sent to your email after your preorder is fulfilled. At Amazon, instead of the gift card, you also have the option of choosing a free pair of Pixel Buds 2a when you preorder. They are our favorite earbuds, but you probably don’t need them if you already have decent headphones.
The Google Pixel 10a isn’t super impressive compared to previous A-series smartphones. In fact, the Pixel 9a is still our favorite Android phone. The two phones are largely similar, even rocking the same chipset. The Pixel 10a does come in some new colors, though, like Fog and Lavender, and the phone is slightly thinner, with a less noticeable camera bump. The screen is a little brighter and a little more scratch-resistant, and the device is made with more recycled materials.
However, there are some software changes in the 10a compared to the 9a. The Pixel 10a is getting some of the same AI features as you’d find in the Pixel 10, plus support for AirDrop as well. The battery lasts up to 30 hours on a single charge, and the phone will have seven years of software and security updates. We can help you decide which Google Pixel to order, and if you decide on the 10a, these gift card deals are definitely worth it (especially if you were already planning on spending money at Amazon or Best Buy anyway).
-
Tech1 week agoA $10K Bounty Awaits Anyone Who Can Hack Ring Cameras to Stop Sharing Data With Amazon
-
Business7 days agoUS Top Court Blocks Trump’s Tariff Orders: Does It Mean Zero Duties For Indian Goods?
-
Fashion7 days agoICE cotton ticks higher on crude oil rally
-
Entertainment7 days agoThe White Lotus” creator Mike White reflects on his time on “Survivor
-
Tech1 week agoDonald Trump Jr.’s Private DC Club Has Mysterious Ties to an Ex-Cop With a Controversial Past
-
Sports7 days agoBrett Favre blasts NFL for no longer appealing to ‘true’ fans: ‘There’s been a slight shift’
-
Business6 days agoEye-popping rise in one year: Betting on just gold and silver for long-term wealth creation? Think again! – The Times of India
-
Fashion1 week agoIndia’s $28 bn reset: How 5 trade deals will reprice its T&A exports

