HomeMachine LearningMachine Learning NewsML Combats Increased Road Issues like Extreme Weather

ML Combats Increased Road Issues like Extreme Weather

The increased frequency of extreme weather events has become impossible to ignore. In 2018, California experienced the most deadly wildfire in its history. Over 85% of Paradise’s town buildings were destroyed as well as neighboring towns. A combination of high wind gusts, drought conditions and overgrown vegetation created a perfect storm to fuel an unprecedented blaze.

With many cities and towns potentially facing similar conditions, it’s up to civic leaders and government officials to begin to look toward preventative measures to help plan for the future. In efforts to be proactive, they try to manage elements within their own control such as trimming trees away from powerlines. But it’s important to also search for and utilize the technology they have access to and help groups identify and quickly respond to these tasks.

One example of a road issue that is monitored is overgrown vegetation. Overgrown vegetation is a significant safety issue that should not be taken lightly. Not only can vegetation fall on power lines and cause fires, but it can also cover critical safety signs, traffic signals and add debris on the road. As a result, the U.S. Department of Transportation’s Federal Highway Administration provides guidelines for local highway and street maintenance to prevent safety hazards.

The process of managing this has typically been a resource-intensive process that requires expensive equipment and boots on the street. Because of the labor cost and safety concerns, an evaluation of road conditions is typically completed once a year. However, managing overgrown vegetation, road debris and other hazards should require a more frequent visual insight of the roads.

Thankfully, there are several technologies that are capable of providing such insights.

Lidar technology is remote optical sensing that is commonly used to detect the elevation of trees and forestation and can be used to estimate the growth of vegetation. The National Oceanic Atmospheric Association (NOAA) uses lidar to make more accurate shoreline maps, digital elevation models and uses it for several other applications. However, lidar can be an expensive and slower option to analyze all the data required.

We live in a time where visual-based machine learning, or computer vision, can analyze an image, recognize, classify, and process it in near real-time, and provide an understanding of what the image is. We see this with facial recognition technology, but in the transportation space, there’s also the option of crowd-sourced images (Nexar is a partner of Blyncsy) and visual map providers that are collected from dash-cam providers, satellites and drones. These images and videos can be curated and run through machine-learning models to quickly identify issues like overgrown vegetation on the roads or other hazards such as misplaced construction barrels on the street. Once these issues are detected, that information can be sent to road maintenance crews so they can immediately address them. This method can streamline the maintenance process while also decreasing response times to issues that are raised, making the roads safer and being proactive on preventative measures needed.

Machine learning models are innovating companies and all industries around the world. This is especially true for the transportation sector. In the context of transportation, algorithms can be run on various forms of data collection like videos and images to create models that will allow for actionable insights to be made. By using machine learning models, organizations that work in transportation can have real-time, automated updates on the conditions of their roadways. Machine learning is used to estimate travel times, predict the best and most efficient route for you to take, and manage traffic flow. It can also take weather data into consideration and port that data right into our preferred map application.

It’s an exciting opportunity to enable more real-time information using lidar technology and machine learning models. This helps make the process more efficient for departments of transportation and make our travel safer.

This article has been published from the source link without modifications to the text. Only the headline has been changed.

Source link

Most Popular