RAI is a multi-vendor agent framework for robotics, utilizing Langchain and ROS 2 tools to perform complex actions, defined scenarios, free interface execution, log summaries, voice interaction and more.
Is your feature request related to a problem? Please describe.
While RAI has an optional voice interactivity package, there is no supporting tool yet to integrate human face & emotion detection, recognition, tracking, pose detection and so on. These are important elements of HRI.
Describe the solution you'd like
It would be best to integrate HRI capabilities in a modular way, respectful of diverse sensor, compute and platform setups. ROS4HRI standard and packages are a good fit to try out.
At first, an RFC for HRI should be proposed to gather community comments and feedback.
Describe alternatives you've considered
Integrating with a paid product, which is fine as an option, but not as a default since HRI is one of important aspects of embodied AI, and relying exclusively on a paid solution would limit accessibility of RAI.
Is your feature request related to a problem? Please describe. While RAI has an optional voice interactivity package, there is no supporting tool yet to integrate human face & emotion detection, recognition, tracking, pose detection and so on. These are important elements of HRI.
Describe the solution you'd like It would be best to integrate HRI capabilities in a modular way, respectful of diverse sensor, compute and platform setups. ROS4HRI standard and packages are a good fit to try out.
At first, an RFC for HRI should be proposed to gather community comments and feedback.
Describe alternatives you've considered Integrating with a paid product, which is fine as an option, but not as a default since HRI is one of important aspects of embodied AI, and relying exclusively on a paid solution would limit accessibility of RAI.
Additional context ROS4HRI github