“Hey robot, navigate to the CNC station and load materials”. Every programmer has imagined it: executing commands with natural language, free from code or modular constraints. Our goal is to turn this vision into reality, and we’re using large language models (LLMs) to push no-code programming to the next level.
Our powerful software platform AgileCore drives this initiative, enabling users to seamlessly plan, execute, and review processes— without having to rely on additional platforms. It also provides a comprehensive skill-library that simplifies programming.
The integration of LLMs has greatly enhanced this ease of use. Large language models are advanced AI systems trained on vast datasets to understand and generate natural language—ChatGPT being one of the most recognized examples. These AI-driven models can communicate naturally, answer questions, and autonomously generate text. Agile Robots utilizes LLMs to simplify robot interaction and enhance process efficiency. To support this, AgileCore offers an extensive range of features.
Step-by-step instructions
When you're struggling to connect an external peripheral device to a robot, the process can quickly become time-consuming, costly, and frustrating. AgileCore's AI assistant offers a solution, streamlining the integration of hardware and software, and enabling seamless robot skill configuration with easy-to-follow, step-by-step guidance. Powered by 'Retrieval Augmented Generation' (RAG) technology, the assistant improves AI accuracy by utilizing data specifically drawn from the AgileCore data pool.
Here’s how it works in practice: if a user needs to calibrate a camera, they can enter a prompt such as “How do I calibrate a camera in AgileCore.” A prompt is essentially a question or instruction that helps the user achieve a specific task. Once the input is confirmed, the AI assistant provides detailed instructions, eliminating the need for additional research. This saves time and enables smoother, more efficient integration.
Commands in natural language
The conversational programming assistant also simplifies the creation of complex motion sequences. Rather than requiring multiple individual commands, users can simply instruct the robot with a phrase like, “Load the CNC machine with the metal block.”
Once the command is issued, the system generates a preconfigured task that the robot executes with precision. It employs advanced vision technology to autonomously recognize both the object and its destination, providing the robot with visual information. This capability also allows for the integration of navigation skills in driverless transport systems, such as autonomous mobile robots (AGVs).
Even after automatic generation by the LLM, it is still possible to make manual adjustments to the parameterization and workflow.
Our AI assistant, instructed by natural language, uses advanced image processing technology, which independently recognizes objects and their target locations. This gives the robot a precise overview of the positions and nature of the objects and allows it to approach and grasp them autonomously.
Intelligent self-optimization
By learning from its mistakes, the robot evolves over time by continuously collecting data. This data is analyzed and evaluated using machine learning algorithms, enabling the robot to automatically make adjustments that enhance its performance.
Additionally, the robot’s control strategies also improve as it leverages these data patterns to autonomously refine its gripping and manipulation techniques. This ongoing improvement enhances the overall process and boosts production efficiency. Moreover, the platform supports making adjustments in real time.
Just like people, our robots never stop learning. By constantly gathering data, they can detect and correct their own errors, enabling them to continually refine workflows with the support of machine learning algorithms
Robotics for everyone
Thanks to the AI assistant, we can further democratize robotics and make it more accessible to a wider audience. Users no longer need programming or robotics skills; they can simply give the robot instructions or ask him questions like they would to a colleague. This support from AI also boosts production efficiency, resulting in substantial time savings and continuous self-optimization that enhance overall productivity.
These innovations represent a new milestone in automation, bringing once-futuristic concepts into the present and ensuring lasting benefits for the fields of robotics and production.
Find out more about our robotic solutions or contact or our experts for a quote.
About the author: Henner Brandes
As a Communications Specialist, Henner Brandes supports the corporate communications of Agile Robots. He is dedicated to presenting robotics topics in a practical, clear, and engaging manner, catering to all levels of expertise. He accomplishes this by collaborating closely with Agile Robots' technical experts. Furthermore, Henner leverages his extensive industry experience, which has been closely connected to robotics and technical communication since the beginning of his career.