Motion Metrics launches LoaderMetrics 2.0

Motion Metrics has dedicated the last year to developing LoaderMetrics 2.0
Motion Metrics launches LoaderMetrics 2.0 Motion Metrics launches LoaderMetrics 2.0 Motion Metrics launches LoaderMetrics 2.0 Motion Metrics launches LoaderMetrics 2.0 Motion Metrics launches LoaderMetrics 2.0

Thermal camera imaging is one of the new features of LoaderMetrics 2.0

Ailbhe Goodbody

The company stated that its LoaderMetrics team has gathered customer feedback worldwide, studied the latest in artificial intelligence and machine learning, and got its hands on the most rugged and durable components to launch its highest performing system to date.

LoaderMetrics is comprised of Missing Tooth Detection and Blind Spot Reduction capabilities. The Missing Tooth Detection function monitors bucket teeth conditions - when a tooth is detected missing, the system alerts the operator and prevents the tooth from reaching and jamming downstream crushers. Blind Spot Reduction features a series of cameras on the loaders’ exteriors, providing real-time surveillance views of their surroundings. In doing so, safety is paramount and dangerous collisions can be avoided.

New features of LoaderMetrics 2.0 include: 

  • Thermal camera imaging;
  • Artificial intelligence and deep learning;
  • System calibration no longer needed; and
  • Lens cleaning system.

Motion Metrics stated that its decision to utilise thermal cameras was made for a myriad of reasons, most importantly to improve system accuracy and precision. Given the short amount of time (less than one second) the system has to capture a clear image of the bucket teeth as the bucket dumps to the truck, a thermal camera focuses solely on bucket teeth and is unaffected by differing backgrounds and the bucket’s contents. Additionally, the system can operate accurately at night without the need for a light, and through dirt and debris on the camera lens and bucket.

LoaderMetrics can already detect missing teeth with high accuracy – but it could do much more if it could perceive, learn and think like a human. To achieve this, Motion Metrics built a deep neural network by inputting hundreds of thousands of real bucket images from mines worldwide. The images are analysed pixel by pixel, detecting patterns and more accurately predicting behaviour. Using the company’s new patent-pending deep learning algorithm, system calibration is no longer needed. Users simply enter the total number of bucket teeth and the algorithm takes care of the rest. As a result, the systems can detect whether a tooth is missing with more accuracy than ever before.

Lastly, the company noted that the inclusion of a lens cleaning system was essential. When compared to its ShovelMetrics cameras, LoaderMetrics cameras are exposed to significantly more debris given their positions beneath the bucket looking upward. The lens cleaning system is integrated into its Motion Metrics embedded systems and uses anti-freeze wash fluid and pressurised air to remove dirt, mud, snow, grime and mist. The lens cleaning system is capable of connecting to a local LAN or any embedded systems through an Ethernet port to be operated remotely.

Motion Metrics expects LoaderMetrics 2.0 to be its highest performing system to date.