eSmart Systems was looking to enhance utility inspections and improve grid efficiency using computer vision for fault detection and predictive maintenance.
By integrating Ultralytics YOLO models into its platform, Grid Vision®, eSmart Systems reduced inspection time by 50%, enabled faster fault detection, and shifted to proactive maintenance.
eSmart Systems is a Norway-based company that helps utility providers inspect and manage large-scale assets, such as power grids and substations, using computer vision and analytics. In particular, their flagship platform, Grid Vision®, leverages computer vision, geospatial analytics, and time-series data to analyze aerial imagery, detect components and defects, and provide predictive insights across transmission lines.
To further enhance inspection efficiency, eSmart Systems integrated Ultralytics YOLO models into Grid Vision®. This resulted in boosted defect detection speed and made it possible for utilities to shift from reactive repairs to more efficient, condition-based maintenance.
With headquarters in Halden, Norway, eSmart Systems focuses on bringing innovative solutions to the utilities sector to monitor and maintain critical infrastructure. For example, their flagship platform, Grid Vision®, provides a comprehensive solution for inspecting and managing large-scale assets like power grids and substations.
Trusted by over 70 utilities worldwide, eSmart Systems has inspected more than 100,000 kilometers of power lines, enabling utilities to make better, data-driven decisions. Grid Vision® makes maintenance more efficient, reduces risks, and supports the transition to more resilient and sustainable energy infrastructure.
eSmart Systems also ensures that its AI solutions meet high standards for data privacy and regulatory compliance. They are ISO 27001 certified for information security management and comply with Netcode Article 7.8, which governs secure data exchange in European electricity grid operations.
Power grids stretch across vast areas, often running through remote or hard-to-reach locations. Many of these systems are aging and require regular inspections to ensure safety and reliability. Inspecting components like transmission towers and power lines is time-consuming, costly, and can be risky for workers.
eSmart Systems aimed to capture aerial imagery using drones and helicopters, applying computer vision to detect components and identify defects. However, because utilities have different components and capture images in various conditions, it was challenging to maintain a consistent inspection workflow.
Manually reviewing these images was also slow and resource-intensive, making it hard to scale fault detection. To automate inspections and support proactive maintenance, eSmart Systems needed a fast and adaptable Vision AI model that could perform reliably across asset types, regions, and weather conditions.
To bring automation and intelligence to grid inspections, eSmart Systems integrated Ultralytics YOLO, a computer vision model, into its Grid Vision® platform. Ultralytics YOLO models support various computer vision tasks, including object detection, which allows the platform to identify key components like towers, crossarms, insulators, and conductors in aerial imagery.
The models are also being used to detect defects such as vegetation encroachment, damage, and wear, which can affect the grid’s performance. Once the components and defects are detected, this information is processed through Grid Vision®, which uses cloud-based processing to automate and scale the inspection process quickly and accurately.
The platform flags potential defects, evaluates the associated risk levels, and helps utilities plan maintenance based on the condition of the assets. This combination of real-time detection and analysis allows utilities to move from reactive maintenance to a more proactive approach, assisting them in staying ahead of potential issues before they lead to costly failures.
By integrating these insights with metadata and time-series data, Grid Vision® enables utilities to optimize their maintenance strategies, improving efficiency and reducing the risk of unexpected outages.
eSmart Systems adopted Ultralytics YOLO models for their speed, accuracy, and seamless integration into their AI pipeline. Ultralytics YOLO models deliver consistent results when analyzing large, high-resolution aerial images, making them ideal for grid inspections.
Also, the Ultralytics Python package offers a variety of integration options, including 15 export formats. This flexibility enables eSmart Systems to deploy the models across different environments. They use formats like PyTorch for training and ONNX for optimized CPU inference in production, particularly when GPU resources are limited in their cloud infrastructure.
With over 30 Ultralytics YOLO models already in production, eSmart Systems can efficiently scale inspections. This allows them to focus on improving data quality and addressing utility-specific challenges.
The impact of Grid Vision®, powered by Ultralytics YOLO models, has been significant in enhancing utility inspections. By automating asset inspections and improving defect detection, Grid Vision® has reduced manual workloads, increased safety, and facilitated more proactive maintenance strategies.
For instance, in Switzerland, a major energy company managing thousands of pylons (tall structures supporting power lines) in mountainous terrain reduced inspection times by 50%. Transitioning from manual climbing to drone-based inspections sped up fault detection, improved worker safety, and saved time.
Similarly, in the United States, a large utility provider used Grid Vision® to digitize 1,400 transmission structures in just three months. This AI-driven image analysis replaced manual photo reviews, allowing for remote validation and enabling better, data-driven capital planning decisions.
Similarly, in Finland, a transmission system operator reduced field visits and minimized outages by switching from ground-based inspections to drone-assisted assessments. With Grid Vision® and YOLO-powered defect detection, inspection accuracy improved, and skilled workers were able to focus on more important tasks.
.webp)
Looking ahead, as eSmart Systems expands globally, they are addressing challenges such as varying infrastructure, different image capture methods, and data drift across regions. To overcome these concerns, the company is focusing on making Grid Vision® more scalable and adaptable.
Their progress with MLOps pipelines has been key, simplifying model retraining and automating dataset expansion. These improvements continuously enhance the accuracy and performance of their AI solutions. eSmart Systems is paving the way for more efficient and reliable grid management, ensuring a future-ready approach to the global energy transition.
Interested in computer vision? Explore our GitHub repository to see how Ultralytics YOLO models are driving innovations in areas such as AI in self-driving cars and computer vision in agriculture. Learn more about our YOLO models and licensing options today!
Ultralytics YOLO models are computer vision architectures developed to analyze visual data from images and video inputs. These models can be trained for tasks including Object detection, classification, pose estimation, tracking and instance segmentation.Ultralytics YOLO models include:
Ultralytics YOLO11 is the latest version of our Computer Vision models. Just like its previous versions, it supports all computer vision tasks that the Vision AI community has come to love about YOLOv8. The new YOLO11, however, comes with greater performance and accuracy, making it a powerful tool and the perfect ally for real-world industry challenges.
The model you choose to use depends on your specific project requirements. It's key to take into account factors like performance, accuracy, and deployment needs. Here's a quick overview:
Ultralytics YOLO repositories, such as YOLOv5 and YOLO11, are distributed under the AGPL-3.0 License by default. This OSI-approved license is designed for students, researchers, and enthusiasts, promoting open collaboration and requiring that any software using AGPL-3.0 components also be open-sourced. While this ensures transparency and fosters innovation, it may not align with commercial use cases.
If your project involves embedding Ultralytics software and AI models into commercial products or services and you wish to bypass the open-source requirements of AGPL-3.0, an Enterprise License is ideal.
Benefits of the Enterprise License include:
To ensure seamless integration and avoid AGPL-3.0 constraints, request an Ultralytics Enterprise License using the form provided. Our team will assist you in tailoring the license to your specific needs.
.png){kind=logo}
.webp)
.webp)
.webp)