Weighing things are getting more complex and advance every day. One of a good example is how to weigh moving trucks. The conventional scales are just not fast and accurate enough. Find out how Artificial Intelligent solves this problem.
Many have use Average-based Calibration Algorithm to calculate the axle weight of trucks. It collects as many weights as possible in a period of time until it is able to get a consistent average weight. Because of this, trucks must move slowly on a Weigh Bridge to get a longer period of time on a weighing scale. Furthermore, the weights collected is not as accurate because the existence of noise and vehicle dynamics. Vehicle dynamics is largely referred to the distance among truck axles.
The solution is to identify the noises and vehicle dynamics, and include them in the calculation. This way we can neutralize the noises and get an accurate weight more quickly. Due to the complexity, the most suitable candidate in solving this would be AI, Artificial Intelligent using Neural Networks. Neural Networks can identify the underlying relationships such as the spatial repeatability in axle dynamics. It can also efficiently identify noises. Furthermore, they can adapt to changing circumstances. Like for example traffic condition, road profile or sensors failure.
The major component of the solution lies in the process of the Neural Network design, the size of the training sample and the length of the training period. The more training, the more intelligent it becomes. Research has confirmed that the Neural Networks approach does resulted in higher accuracy than the traditional Average-based Calibration method. It also able to work out the weight with less weight points therefore faster result. So, it is a very promising idea to incorporate this in Weigh Bridge Software or Weighing Software.
by Jerry Craft