When designing Robot Freedom, our educational presentation on robotics, the HipMonsters team wanted to make robotics and artificial intelligence (AI) approachable to a mass audience in hopes of inspiring the creators within all of us. To achieve this, the core principles for our AI design were defined by the Hip Monster’s sister team (ages 9 and 12 at the time), namely, robots should have distinct personalities, emotions, curiosity and be first and foremost pieces of art.
Given these principles, the foundation of our artificial intelligence framework (show above) is based on Stimulus Organism Response (S-O-R) Theory. S-O-R theory is a psychological framework that enables researchers to explore how stimuli (such as a bell) can impact an organism’s responses, (a dog salivating). Like Pavlov’s dog salivating at the sound of a bell, our robots learn and adapt as they experience outside stimuli and are always eager for more. The robot’s AI is driven by five personality traits that govern how they interpret and respond to stimuli. Below is how a signal from a sensor (stimuli) flows through our AI (organism) and results in an action (response).
Central to the robot’s stimuli exploration is a sensor array of ten sensors ranging from sound to touch. When a robot receives a stimulus, it first processes the information based on its preset personality, then uses past experiences to choose a response based on its personality. Below is a color key to the robot’s sensor display panel.
These experiences are weighted based on the outcome of the robot’s actions allowing the robot to adapt responses to new stimuli. The robots can move, change visual effects, or talk using a chatbot. Below is the full software stack used in our robots.
All the processing is run on a Raspberry Pi and you can download if on our GitHub. Come see this in action at this year’s Bay Area Maker’s Faire!
Happy creating!