As more bikes than ever take to the city streets, researchers have designed Ari, a smart e-bike that could help riders cruise the ‘green wave’ while also improving trust between humans and machines.
The frustration of being stuck at a red light is universal and electric bike, or ‘e-bike’ riders are no exception. But collaborative research between IBM Research-Australia and RMIT University’s Exertion Games Lab has resulted in a system that allowed humans and bikes to work together to catch more green lights.
Partners in riding
Ari used artificial intelligence (AI) and the Internet of Things to assist the rider to regulate the speed and cross traffic lights on green. Using traffic data and green wave modeling provided by Victoria’s road and traffic authority, VicRoads, we trialed the e-bike in real traffic conditions on Melbourne’s streets.
To achieve this, we calculated the optimum speed required to have the greatest chance of crossing all lights on green at the test location — which happens to be 22km per hour. We then programmed Ari to assist the rider to meet this reference speed. The e-bike assisted the rider physically by increasing engine support to accelerate, and cognitively by whispering “slow down a little” via bone-conducting headphones, which left the rider’s ears uncovered in order to safely navigate the environment.
The project is an example of how advances in AI and the Internet of Things in everyday objects could have implications in many aspects of our day-to-day lives.
Through this research, we can begin to explore what it means for humans to partner with AI systems. And how we can design these systems to support user interaction by being trustworthy, explainable, and, ultimately, explore futures that could make a societal contribution.
This research is also important as it explores a new type of interaction between human and computer (and machine), where the computer does not replace the user’s exertion (i.e. it is not a motorbike that replaces pedaling), but instead it can physically and cognitively support the rider, offering opportunities for partnership.
Computer says coach
Real-world applications of the e-bike technology could depend on per-country traffic light data access. Currently, there are no plans for exploration. However, what we can do now with the learnings from this experiment is inform the design of the user and AI system partnership, specifically in contexts where screens like smartphones are not needed, as they can be an obstruction to the user while moving.
We believe the development of reliable and explainable AI could open opportunities where this type of systems can serve as human “coaches” by complementing physical effort and offering extra cognitive abilities.
A team of researchers from IBM Research AI and AI Horizons Network-partner the University of Michigan published the papers “A Large-Scale Corpus for Conversation Disentanglement” and “Learning End-to-End Goal-Oriented Dialog with Maximal User Task Success and Minimal Human Agent Use” at ACL 2019. This work address two main challenges in building enterprise AI assistants.
The latest work on computational argumentation from the IBM Project Debater research team group is being presented at the ACL 2019 conference. Three papers will be presented at the main conference and one more paper will be presented in the co-located Argument Mining Workshop.