Big tech quest to find essential metals for energy overhaul


One reason the company focused its initial efforts on Canada is that the nation has a large amount of publicly available survey data, including narrative field reports, obsolete geological maps, geochemical data on borehole samples, aerial magnetic and electromagnetic survey data. , lidar readings, and satellite imagery spanning decades of exploration.

“We have a system where we can take all that data and store it in standard formats, quality check all the data, make it searchable, and access it programmatically,” Goldman says.

Hi-tech momentum

After compiling all available information for a site, KoBold’s team explores the data using machine learning. For example, the company could build a model to predict which parts of ore deposits have the highest concentration of cobalt, or create a new geological map of an area that shows all the different rock types and fault structures. Goldman could add new data to these models as it is collected, allowing KoBold to adaptively change its exploration strategy “in near real time.”

Canada makes datasets and other information publicly available, such as this lidar-generated image of Saskatchewan.


KoBold has already used information from machine learning models to derive Canadian mining claims and develop field programs. Partnership with Stanford’s World Resources Estimation Center, which has been underway since February, adds an additional layer of analytics to the mix in the form of an AI “decision agent” that can map out an entire exploration plan.

Stanford geoscientist Jef Caers, who oversaw the collaboration, explains that this digital decision maker measured the uncertainty in KoBold’s model results and then devised a data collection plan to sequentially reduce that uncertainty. Like a chess player trying to win a game in the fewest moves possible, the AI ​​will aim to help KoBold reach a possible decision with minimal wasted effort – whether it’s punching a certain point or walking away.


Source link

Leave a Reply

Your email address will not be published. Required fields are marked *