Rekor Manages Traffic using NVIDIA AI

As more people move to Texas, Rekor uses NVIDIA technology to improve roadway safety and alleviate traffic.

The business is using AI-driven analytics with NVIDIA AI, Metropolis, and Jetson for roads in Texas and Philadelphia, which might reduce fatalities and improve quality of life.

Austin is drawing tourists with jobs, comedy clubs, music venues, barbecues, and more. However, one of the biggest city blues that has accompanied this expansion is traffic congestion.

Rekor, which offers traffic management and public safety analytics, has a clear perspective of the increasing traffic because of the influx of new residents to Austin. Rekor works with the Texas Department of Transportation, which is pursuing a $7 billion program to address the roadway problems.

With its headquarters located in Columbia, Maryland, Rekor has been using NVIDIA Metropolis for real-time video comprehension and NVIDIA Jetson Xavier NX modules for edge AI in Texas, Florida, Philadelphia, Georgia, Nevada, Oklahoma, and many other U.S. regions, as well as Israel and other nations.

A vision AI application framework for building intelligent infrastructure is called Metropolis. The NGC catalog pretrained models, TensorRT, TAO Toolkit, and NVIDIA DeepStream SDK are some of its development tools. For embedded and robotics applications, the small, potent, and energy-efficient NVIDIA Jetson accelerated computing platform is perfect.

The latest development in a lengthy history of traffic management and safety is Rekor's efforts in Texas and Philadelphia to use AI to enhance road management.

Minimizing Jams, Fatalities, Pileups, and Rubbernecking

Rekor offers two main products: Rekor Command and Rekor Discover. Command, an AI-driven program, enables traffic control centers to promptly detect traffic accidents and areas of concern. Transportation authorities can manage safer and less clogged municipal roadways thanks to the real-time situational knowledge and alerts it gives them.

Discover fully automates the gathering of comprehensive vehicle and traffic data using Rekor's cutting-edge technology. It also provides robust traffic analytics that convert road data into measurable, reliable traffic statistics. Rekor Discover provides transportation departments a thorough understanding of how automobiles move on roads and the impact they have, which may help them plan and execute their future city-building initiatives more effectively.

The company has distributed command around Austin to help with incident analysis, issue identification, and real-time traffic response.

Weather, connected vehicle information, traffic camera footage, construction updates, and data from other parties are just a few of the data sources that Rekor Command gets. It then uses AI to create connections and identify anomalies, like a roadside incident. The data is sent to traffic control centers for analysis, confirmation, and response.

Rekor is using NVIDIA's full-stack accelerated computing for highway intelligence as part of the NVIDIA AI Enterprise software platform. It is also making significant investments in NVIDIA AI and NVIDIA AI Blueprints, standard workflows for generative AI use cases built with NVIDIA NIM microservices. A group of intuitive inference microservices called NVIDIA NIM was created to expedite the installation of foundation models on any cloud or data center while preserving data security.

Using the NVIDIA AI Blueprint for video search and summarization, Rekor is creating AI agents for municipal services, namely in areas like infrastructure optimization, traffic management, and public safety. NVIDIA has unveiled a new AI architecture for video search and summarization that will allow for a range of interactive visual AI agents capable of extracting complex behaviors from large volumes of recorded or live video.

Philadelphia Keeps an Eye on Pollution, EV Charger Needs, and Roads

Managing the roadways and gathering data on new development are challenges for the Philadelphia Industrial Development Corporation (PIDC), which is in charge of the Philadelphia Navy Yard, a well-known tourist attraction. The Navy Yard site will generate 12,000 jobs and thousands of residents with more than 150 businesses and 15,000 employees spread over 1,200 acres, according to a $6 billion redevelopment plan.

PIDC aimed to increase awareness of how construction projects and road closures affect mobility as well as strategies for enhancing mobility during significant events and projects. PIDC also aimed to increase the Navy Yard's ability to understand the quantity and movement of car carriers and other heavy vehicles, as well as monitor the impact of speed-mitigating devices positioned across hazardous road segments.

Discover provided PIDC with information on additional infrastructure projects that need to be put in place to manage any traffic variations.

PIDC can make well-informed judgments regarding future EV charging station sites by tracking the number of electric vehicles entering and leaving the Navy Yard. Use of Rekor Discover, which collects data from Rekor's edge systems built with NVIDIA Jetson Xavier NX modules for powerful edge processing and AI to understand the number of EVs and where they're entering and departing, will help Navy Yard better plan potential locations for EV charge station deployment in the future.

PIDC planners were able to create a hotspot map of EV traffic by using Rekor Discover to analyze data provided by the AI platform. For real-time traffic analysis, the system makes use of Jetson and NVIDIA's DeepStream data pipeline. It utilizes the NVIDIA Triton Inference Server to enhance LLM capabilities even further.

In addition to addressing public safety issues over speeding and collisions, the PIDC aimed to minimize property damage. Speed insights are being used to execute traffic calming measures when average speeds on certain road segments exceed the approved speed limit.

Real-Time Pollution Monitoring with NVIDIA Jetson Xavier NX

Compared to the traditional approach of utilizing satellite data to try to understand its locations, Rekor's vehicle recognition models, which were powered by NVIDIA Jetson Xavier NX modules, were able to track pollution to its sources, bringing it one step closer to abatement.

Rekor is looking at how NVIDIA Omniverse may be used in the future to build digital twins that simulate traffic reduction using different methods. Omniverse is a platform for developing OpenUSD applications for industrial digitalization and generative physical AI.

Utilizing Omniverse to create digital twins for communities has important implications for reducing traffic, pollution, and traffic fatalities all of which Rekor considers very beneficial for its customers.