AMD Silo AI has announced a partnership with Robotec.ai, a developer specializing in simulation platforms for robotics applications. This collaboration aims to optimize and scale digital twin and scenario reconstruction workloads, specifically for next-generation automotive and robotics systems utilizing AMD Instinct GPUs and the ROCm software stack.
The joint effort aligns with Robotec.ai’s mission to foster safe, human-friendly robotics. Through this partnership, AMD Silo AI will leverage Robotec.ai’s open-source, AI-driven digital twin simulation tools. The collaboration builds on Robotec.ai’s ongoing work with AMD Kria SOM, combining their flagship RoSi platform—an end-to-end tool for developing, training, and testing robotics and autonomous systems—with AMD’s scalable compute infrastructure.
A notable feature of the RoSi platform is its RoSi Sensors, which utilize the RGS ray-tracing library. This technology enhances GPU performance for LiDAR and radar simulations, allowing for rapid testing of intricate multi-sensor configurations.
Focus on Open-Source Robotics Solutions
According to Michal Niezgoda, CEO and co-founder of Robotec.ai, “Open source empowers customers to evaluate technology on their own terms, inspect code for cybersecurity, and benefit from a community that validates and improves solutions.” He emphasized that reliance on proprietary solutions often limits flexibility and innovation.
The partnership aims to expand the robotics ecosystem by developing open-source solutions and accelerating development cycles. It also opens opportunities for academic collaborations and contributions from third parties. Peter Sarlin, CEO and co-founder of AMD Silo AI, highlighted that this collaboration reflects a mutual commitment to open-source innovation. “This commitment empowers customers with digital sovereignty and seamless portability of GPU workloads—enabling safer, more efficient robotics systems across diverse hardware and environments,” Sarlin explained.
Advancing Autonomous Systems and Robotics
The partnership is part of a broader initiative by AMD Silo AI to establish an open ecosystem for autonomous systems and robotics. This initiative also includes a previously announced collaboration with Parallel Domain, demonstrating AMD’s commitment to integrating top-tier software with cutting-edge GPU infrastructure in the autonomous and robotics sectors.
One of the key projects under this collaboration involves testing the capabilities of the first fully autonomous warehouse robot. This robot uses agentic AI to dynamically plan and execute tasks in real-time, eliminating reliance on hard-coded scripts. It operates solely on AMD Ryzen AI processors designed for embedded systems.
By leveraging its extensive experience in end-to-end computing solutions—including safety-critical edge devices, digital twin platforms, and AI cloud infrastructure—AMD Silo AI aims to broaden access to high-performance robotics simulation on a global scale. The Finnish AI lab was acquired by AMD last year, marking a significant step in the company’s efforts to enhance its capabilities in the rapidly evolving field of robotics.
