This week, we’re celebrating the more than 2 million developers now using the NVIDIA robotics stack. These builders are reshaping industries across manufacturing, food delivery, agriculture, healthcare, facilities maintenance and much more.

Since the launch of the NVIDIA Jetson platform in 2014, a growing ecosystem of more than 1,000 hardware systems, software and sensor partners have joined the thriving developer community to help enable more than 7,000 customers to adopt edge AI across industries.

The next-generation NVIDIA Jetson Thor platform is built for physical AI and humanoid robotics. It supports any popular AI framework and generative AI model and is fully compatible with NVIDIA’s software stack from cloud to edge, including NVIDIA Isaac for robotics simulation and development, Isaac GR00T humanoid robot foundation models, NVIDIA Metropolis for vision AI and NVIDIA Holoscan for real-time sensor processing.

There’s a staggering number of industry-changing applications in the market enabled by NVIDIA Jetson. We’ll be featuring just some of the companies here daily all week, and stay tuned: some exciting product news is coming on the last day.

Read about:

• Agriculture — Carbon Robotics
• Delivery — Serve Robotics
• Manufacturing — Peer Robotics

Laser-Focused: Carbon Robotics Uses NVIDIA GPUs to Burn Weeds and Break New Ground (Literally) 🔗

In farming, where work is still measured by acres and hours in the sun, a new kind of tool is cutting through one of agriculture’s oldest problems: weeds.

Carbon Robotics’ story began not in a lab, but over lunch in Idaho. Founder Paul Mikesell asked a farmer about his biggest challenges. The answer was simple — and ancient: weed control.

That conversation led to an audacious solution. The company’s LaserWeeder G2 spans 20 feet and carries 12 modules, each powered by two NVIDIA GPUs — 24 in total. This compute muscle lets it identify and incinerate up to 10,000 weeds per minute, without chemicals.

In an age of herbicide-resistant plants, it’s a game-changer. “There’s no such thing as a laser-resistant weed,” says Alex Sergeev, chief technology officer of Carbon Robotics.

Every LaserWeeder doubles as a data-gathering machine. Images from the field feed into a custom labeling tool, building what the team believes is the world’s largest labeled agricultural image dataset: over 65 million images, fueling the company’s “large plant model,” a play on the large language models that power chatbots like ChatGPT.

When GPUs finish their tour on the farm, they head to Carbon’s “retirement facility” — a Seattle-based data center — where they’re used to train the next generation of models. Optimized with NVIDIA TensorRT, the resulting foundational model works across all crops, creating a durable competitive edge.

From there, Carbon turned to another urgent challenge: a global shortage of tractor drivers. Over 25% of edible U.S. crops go unharvested due to labor gaps. The answer: the Carbon AutoTractor, an autonomous retrofit for existing machines. Built with farmers’ input, it runs around the clock, monitored by a remote operations center that can take control if, say, a deer wanders into the field.

Since its founding in 2018, Carbon Robotics has shipped more than 150 LaserWeeders to farmers in 14 countries, eliminating over 30 billion weeds. With investment from NVentures, NVIDIA’s venture arm, this Pacific Northwest company is proving that physical AI can tackle some of humanity’s oldest problems — from the ground up.

Serve Robotics Is Ushering Food Delivery Into the Mainstream 🔗

Serve Robotics is scaling up its fleet of food delivery robots to 2,000 by year’s end, entering Chicago and expanding across Los Angeles, Miami, Dallas-Fort Worth and Atlanta. Delivery partners include Uber Eats, Shake Shack, Little Caesars and 7-Eleven.

Serve, which went public last year, runs its delivery robots with three NVIDIA GPUs—including one Orin AGX. It needs a lot of real-time processing to handle data from multiple cameras and lidar for navigation and obstacle detection, says co-founder and CEO Ali Kashani.

“We plan to quadruple the number of robots we have and launch at least two cities by the end of the year. So we have been expanding quite significantly and very quickly,” said Kashani. “Right now, about 1.8 million people can use our service across the country.”

Rolling through crowded streets among people for years, Serve’s delivery robots rely on robust simulation models developed with the NVIDIA Isaac SDK.

Serve recently acquired Vayu Robotics, a startup specialized in an end-to-end foundation model-based approach to navigation and autonomy. Integrating Vayu’s innovations in simulation-based AI training with Serve’s leading autonomy platform and growing real-world dataset is expected to accelerate the company’s autonomy roadmap and enable even safer, more intelligent navigation.

“When you’re out in the real world and you see those edge cases, It’s really valuable to be able to model those and then try them again in simulation and test other scenarios — a lot of testing and validation happens in Isaac Sim,” said Kashani.

As one of the most visible robot makers, Serve is also seeing its robots become pop culture stars. Their IMDB credits include the latest Beverly Hills Cop movie and an upcoming movie starring Seth Rogen and Keanu Reeves.

If you’re ready for home movie night with pizza delivery, Serve Robotics will deliver a slice of entertainment of its own to your doorstep.

Peer Robotics Develops Autonomous Pallet Jacks for Manufacturers 🔗

Peer Robotics is rolling into developments with a leading retailer and a large automaker.

Peer Robotics, an NVIDIA Inception startup, deploys NVIDIA Jetson for its lines of manufacturing and logistics robots. Peer recently released its Peer 3000, an autonomous pallet jack assistant that runs on Jetson AGX Orin.

“NVIDIA Jetson has allowed us to create robots that run models for processing camera inputs to safely navigate surroundings and assist workers in doing their work more efficiently, driving better outcomes for businesses,” said Rishabh Agarwal, cofounder and CEO of Peer Robotics. “As a result, our business at Peer Robotics is growing exponentially.”

The Peer 3000 can move payloads up to 3,000 lbs and act as a locomotive to automate complete operations of moving existing trolleys or pallets in warehouses or production lines.

The company processes large quantities of sensor data at the edge with Jetson AGX Orin as well as on the cloud and in local systems — and it relies on a three computer solution for building physical AI for robotics.

Peer Robotics harnesses Isaac Sim with data its robots collect to refine how they can scoop up palettes, pull carts and navigate better in changing warehouse scenarios.

“The Isaac SDK and Isaac Sim have helped us test and validate in simulation before deployment to get to market faster when it comes to handling new pallet types or trolleys and different environment settings,” said Agarwal.

The company’s Peer Robotics Dashboard enables facilities managers to make more informed decisions driven by data collected by its robots.

The company is anticipating the availability of NVIDIA Jetson Thor to run more robust models, handle privacy on the edge and tackle new areas of automation and tasks.