Blogs

AI Explained by ASTERRA’s Lead Data Scientist

March 15, 2022
ASTERRA Profile Photo
By ASTERRA

The use of artificial intelligence (AI) by ASTERRA goes a long way toward simplifying water leak detection and infrastructure maintenance in communities across the globe.

Inon Sharony is head of AI and leads ASTERRA’s data science team, working closely with Yuval Lorig, ASTERRA’s vice president of research and development. A computational scientist in chemical physics, Inon has a deep resume of developing cutting-edge technology using data science.

We caught up with Inon recently to cut through the hype often found in the media related to AI and to learn more about how ASTERRA is using science to support data-driven solutions to real world problems.

AI is a general term for any tools representing computer-aided decision-making. Inon said, “It refers to any machine or device that manifests something we perceive as intelligence.”

Capable of perception, logic, and learning, AI crosses disciplines, including computer science, math, biology, psychology, philosophy, and sociology. There are countless fields, including robotic technology, cloud computing, big data, and decision support.

Under the AI umbrella is machine learning, deep learning, and data science. Machine learning utilizes methods called statistical inference algorithms that learn from data to make predictions or decisions without explicit programming. It takes one big data set and trains itself on half of it; then using the knowledge it learns from the data, it looks through the second half to make educated inferences.

In the machine learning process, performance improves when exposed to more data over time. Among the benefits of using Machine Learning for modelling is the knowledge gained by studying these models.

“Machine learning systems, by and large, exhibit a higher degree of generalization,” Inon said. The technology evaluates all variables to make its predictions. “There may be very different causes for variation. We treat everything with equal seriousness.”

ASTERRA Recover uses satellite imagery and the power of AI to cover large areas and quickly narrow down the regions that contain probable water leaks.

How does ASTERRA do this?

Specifically, L-band synthetic aperture radar (SAR) sensors are used for their day/night, cloudy/clear capabilities, along with the ability to penetrate the first few meters of earth. Using a patented algorithm and machine learning, Recover filters out the signature of drinking water for the customer. The locations are then provided as a GIS data project directly to the utility’s preferred field crew to search within the zones and pinpoint the exact leak location.

Since ASTERRA’s process is one that can occur anytime and more quickly than traditional boots on the ground and acoustic methods, ASTERRA simplifies leak detection and infrastructure maintenance, yielding more energy efficient and sustainable programs.

The use of artificial intelligence (AI) by ASTERRA goes a long way toward simplifying water leak detection and infrastructure maintenance in communities across the globe.

Inon Sharony is head of AI and leads ASTERRA’s data science team, working closely with Yuval Lorig, ASTERRA’s vice president of research and development. A computational scientist in chemical physics, Inon has a deep resume of developing cutting-edge technology using data science.

We caught up with Inon recently to cut through the hype often found in the media related to AI and to learn more about how ASTERRA is using science to support data-driven solutions to real world problems.

AI is a general term for any tools representing computer-aided decision-making. Inon said, “It refers to any machine or device that manifests something we perceive as intelligence.”

Capable of perception, logic, and learning, AI crosses disciplines, including computer science, math, biology, psychology, philosophy, and sociology. There are countless fields, including robotic technology, cloud computing, big data, and decision support.

Under the AI umbrella is machine learning, deep learning, and data science. Machine learning utilizes methods called statistical inference algorithms that learn from data to make predictions or decisions without explicit programming. It takes one big data set and trains itself on half of it; then using the knowledge it learns from the data, it looks through the second half to make educated inferences.

In the machine learning process, performance improves when exposed to more data over time. Among the benefits of using Machine Learning for modelling is the knowledge gained by studying these models.

“Machine learning systems, by and large, exhibit a higher degree of generalization,” Inon said. The technology evaluates all variables to make its predictions. “There may be very different causes for variation. We treat everything with equal seriousness.”

ASTERRA Recover uses satellite imagery and the power of AI to cover large areas and quickly narrow down the regions that contain probable water leaks.

How does ASTERRA do this?

Specifically, L-band synthetic aperture radar (SAR) sensors are used for their day/night, cloudy/clear capabilities, along with the ability to penetrate the first few meters of earth. Using a patented algorithm and machine learning, Recover filters out the signature of drinking water for the customer. The locations are then provided as a GIS data project directly to the utility’s preferred field crew to search within the zones and pinpoint the exact leak location.

Since ASTERRA’s process is one that can occur anytime and more quickly than traditional boots on the ground and acoustic methods, ASTERRA simplifies leak detection and infrastructure maintenance, yielding more energy efficient and sustainable programs.

Please contact us to learn more about ASTERRA products and services.

  • This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
  • This field is for validation purposes and should be left unchanged.