MODEX 2024: RGo Robotics to partner with Intel and Theobroma Systems

Machine vision software provider RGo Robotics and computer hardware provider Theobroma Systems recently announced availability of an integrated computer system for mobile robots. The system combines RGo’s Perception Engine with Theobroma’s JAGUAR computer board.

RGo also announced at MODEX 2024, in collaboration with Intel, that its Perception Engine is optimized to run on Intel core processors. The trade show was held March 11-14 at the Georgia World Congress Center in Atlanta, GA. 

The RGo team will also be available at LogiMAT in Stuttgart, Germany from March 19-21 in Hall 6, stand D25.

Theobroma Systems JAGUAR designed for mobile robots

RGo’s integrated system is based on Theobroma’s new JAGUAR SBC-RK3588-AMR single-board computer. Theobroma said its JAGUAR board’s connectivity, real-time processing power, memory, and energy efficiency have been optimized for robotics applications powered by AI and visual perception.

“The hardware is designed for industrial mobile robots and has all the functionality and connectors needed for the entirety of the robotic solution’s lifecycle,” said Amir Bousani, RGo CEO and co-founder. “Its reliability, affordability and ease of integration makes the solution an appealing option for robot manufacturers when considering whether to use visual perception.” 

Theobroma’s JAGUAR board incorporates Rockchip’s RK3588 SoC with an eight-core CPU, a Mali G610 GPU, and an NPU with up to 6 TOPs with support for up to four MIPI CSI cameras.

RGo Perception Engine can function in dynamic environments

RGo’s Perception Engine is fully integrated into the JAGUAR board. It can track robot pose, location, and heading to centimeter-scale accuracy. RGo said its machine vision software can function in dynamic, unstructured, or repetitive environments, indoors and outdoors, and on fast-moving machines such as forklifts and cart tuggers. 

The companies said their system can be integrated into a wide range of industrial robots such as autonomous mobile robots (AMRs) and automated guided vehicles (AGVs), as well as professional service robots such as those used in transportation, hospitality and agriculture.

“The joint solution offers a perfect balance between performance and cost, enabling manufacturers of robots and industrial automation machines to rapidly open up new business areas with innovative products,” said Stefan Roth, managing director at Theobroma Systems.

Intel processors to support AI-powered RGo Perception Engine

RGo’s Perception Engine offers robot manufacturers AI-powered perception with localization and obstacle detection modules, which the company said allows mobile robot providers to achieve robustness, reliability, speed, and accuracy. RGo said its software enables systems integrators to simplify and shorten new site installations, to operate effectively in dynamic and very repetitive environments, and to improve their spatial understanding, operating in real time with limited latency.

RGo said its Perception Engine optimized for Intel architecture uses different components and embedded hardware accelerators within Intel architecture. Several features include:

• Minimized system load and power consumption: the software assigns each task to the most suitable hardware module - CPU, GPU, and other accelerators - minimizing the system load and power consumption overall.
• Visual data processing: efficient use of resources allows more visual data to be processed, and more sophisticated and complex algorithms to run on Perception Engine, compared to processors with equivalent power and cost, said RGo. In turn, this processing power enables better reliability, higher accuracy, lower latency and more advanced features to be available.
• Single compute platform: with reduced resource consumption, the full robot software stack can run on a single computing platform. 

“Robotic solution providers deploying RGo's Perception Engine can now benefit from a solution that's been optimized for Intel Core processors to help deliver increased performance, reduced power and improved efficiency in automation,” said Sunita Shenoy, senior director of technology products in the federal and industrial solutions division at Intel.

Intel compatibility to facilitate machine vision upgrades without rewriting codebase

This new release of Perception Engine will enable robotics and automation solution providers to keep their Intel architecture-based products and upgrade the visual perception capabilities of their robots without needing to re-write their existing codebase.

RGo said Intel’s recent support for direct camera connection over MIPI enables building simplified and cost-efficient robotics platforms with visual input while utilizing Intel’s powerful platforms for advanced AI applications.

“Providers of robotic solutions often seek to upgrade their perception capabilities. Many are satisfied with their current Intel-based Industrial PCs and wish to consolidate all their software onto the same architecture,” said Bousani. “By providing breakthrough Perception Engine capabilities to robotic platforms powered by Intel IPCs, we enable them to save time, resources, and money that would otherwise be spent on replacing their hardware platforms.” 

Check out Robotics 24/7’s Wednesday MODEX 2024 recap for more information about RGo Robotics.