This talk deals with the usage of artificial intelligent techniques in humanoid robotics. The focus is on human–robot interaction with the goal of building a robotic platform which embodiments are able to interact in natural and effective way with humans through speech, gestures, and facial expressions. The system is fully implemented in Python and based on the Robotic Operating System (ROS). The talk will describe the hardware and software configuration of our currently NAO based humanoid platform. The strategy has been to use available high level Python libraries for spoken language processing, sentiment analysis, vision, interfacing with Artificial Intelligence applications in order to provide current edge technologies performances. The overall system architecture is based on finite state machines nodes interacting via the ROS communication layer.
The main fields of applications that the platform is targeting are:
- Field robotics
- Home and companion robotics
- Robot Assisted Therapy (RAT)
We will present the latest status of the platform together with a NAO based demo.