I tested the early access version and the core idea is solid. Instead of relying on ICL or RAG, queries are translated into a DSL that returns the answer directly. I expect this approach to become the standard for NL Q&A moving forward. The workflow still needs refinement and a few steps are manual, but that can all be solved with a bit of engineering work.
I'm a Data Scientist, Software Engineer, and Robotics Integration Lead at NASA Jet Propulsion Laboratory. Over the past year, our team has been working on the development of an AI agent that is fluent in ROS and can be used by developers of all skill levels.
ROSA (ROS Agent) is augmented with all manner of tools, including ROS1 and ROS2 tools, math tools, and more. It can also be adapted to new robots with unique capabilities by simply augmenting the core agent with new tools and system prompts. (see Developer Docs and Custom Agents on our GH Wiki).
Simply type in your query and ROSA will use its tools to satisfy the query. Some example queries include:
- "Show me a list of topics that have publishers but no subscribers."
- "Set the `/velocity` param to `1.5`
- "Echo the `/robot/status` topic"
Additionally, we added a quick and easy way to demo the agents capabilities by creating a custom agent for the TurtleSim bot. This agent serves as both an easy-to-use demo (in Docker), and as a how-to guide for creating your own custom agents.
As for the practical benefits of ROSA: we did several evaluations in which we asked human experts the same questions we asked of ROSA. For instance, "give me a list of 'status' topics, along with the message type, publishers, and subscribers, in a table format." The results were stunning, with cost savings anywhere between 5000% to 25,000%.
In the future, we plan to release agents for JPL's Open Source Rover and Boston Dynamics Spot robot, to name a few.
We would love to get your opinions and we are open to contributions from the community!