Train TJBot to Wave with an Intent in Node-RED

Today we’ll train TJBot how to understand natural language and wave the arm using the converse node and the Watson Conversation service.

TJBot can use intents, entities, and dialog responses from the IBM Watson Conversation service to understand natural language and take action as requested. Create a Conversation service to train the intent #wave. Feel free to add other phrases to recognize.

And train an entity, @position, with the positions shown.

In the Dialog tab, check for the condition when #wave and @position are recognized and respond with a confirmation of what motion was recognized.

Copy the Conversation credentials into the Conversation section in the TJBot configuration.

Make sure to enable the servo by ticking the servo option in the TJBot configuration.

When the wave intent is recognized, TJBot outputs the confirmation that it has waved.

When the wave intent and position entity is recognized, TJBot outputs the confirmation that it has performed the requested motion.

Here’s a video of how to train TJBot to recognize the wave intent and position entity using the IBM Watson Conversation service.

That’s it for today’s skill. What can you train TJBot to do now that it can wave its arm using natural language commands?

Come back tomorrow and we’ll train TJBot to be a translator.

This post is part of a series of skills you can train TJBot to perform.