IBM Cognitive Technologies Provide Solar Forecasting for University of Michigan Solar Car in the World Solar Challenge

IBM Research announced that it will provide its advanced solar‑forecasting technology to the University of Michigan (UM) team for the 2015 Bridgestone World Solar Challenge, an 1,800‑mile race across the Australian Outback.

The UM team will use IBM’s cognitive computing expertise to obtain real‑time insights into cloud cover, wind patterns, and solar power availability, helping them decide how to drive their solar‑powered car more efficiently.

IBM’s approach blends data from sensor networks, local weather stations, sky‑camera observations, satellite data, and multiple weather models, using machine learning to create a super‑model that is up to 30 % more accurate than conventional forecasts.

The technology was developed under the U.S. Department of Energy’s SunShot Initiative, aiming to improve solar‑forecast accuracy for better integration of renewable energy into the national grid.

According to Leda Daehler, chief strategist of the UM Solar Car Team, accurate solar‑radiation forecasts are crucial for race strategy, and IBM’s technology provides a significant advantage.

The collaboration includes two forecasting techniques: a multi‑day forecast that predicts solar energy availability along the race route, and a near‑real‑time sky‑camera system mounted on scout cars that maps cloud location and transparency to optimize speed, potentially saving up to 15 minutes per day.

Pavan Naik, program manager for the UM Solar Car Team, highlighted that the new data will allow the team to know where clouds are and adjust speed to “chase the sun.”

IBM scientists benefit from the partnership by testing and refining their forecasting methods in a unique, high‑stakes environment, advancing research that can impact many weather‑sensitive industries.

Dario Gil, IBM Research’s vice president of Science and Technology, emphasized that combining physical analytics with data analytics and machine learning drives innovations that can transform multiple sectors.

For more information and live updates on the UM Solar Car Team’s preparation and race, visit http://ibm.co/solarforecasting and http://umicheng.in/solar25.