Great search for technology to find the metals needed to revise energy
[ad_1]
Part of the reason the company has focused its initial efforts on Canada is that the nation has a wealth of survey data in the public domain, such as reports of narrative areas, time-lapse geological maps, geochemical data on drill samples, aerial magnetic and electromagnetic surveys. , lidar readings and satellite imagery over many decades of exploration.
“We have a system in place to swallow all of this data and store it in standard formats, to be able to control all the data, perform searches, and access the program,” Goldman says.
High-tech boost
After compiling all the information available for a site, KoBold’s team analyzes the data using machine learning. The company, for example, can build a model to predict which parts of the mineral deposits have the highest cobalt concentrations or create a new geological map of a region that shows all rock types and fault structures. As it collects, it can add new data to these models, allowing KoBold to change its exploration strategy appropriately “in near real time,” Goldman says.
KoBold has already used machine learning models to acquire Canadian mining claims and to develop programs in its field. His collaboration with Stanford Land Resource Forecast CenterIt adds an additional layer of analysis that has been underway since February to the mix, which can map an entire exploration plan as an AI “decision agent”.
Stanford scientist Jef Caers, who is overseeing the collaboration, explained that this digital decision quantifies uncertainty in the results of the KoBolden model and then designs a data collection plan to sequentially reduce that uncertainty. Like a chess player who wants to win a game in as few moves as possible, AI will aim to help KoBold with a minimum of unnecessary effort — whether the decision is to punch in or out of a particular place.
[ad_2]
Source link