Simple and Effective Load Volume Estimation in Moving Trucks using LiDARs

Industries need to track the amount of materials and goods transported through processing units in order to optimize production. In large-scale industries, trucks and trains are commonly used for transportation. The manual evaluation of the volume of material being transported by these vehicles can be imprecise, inefficient, and even unsafe for employees. Therefore, this work presents an automated system for estimating the volume of load in moving trucks using a pair of multi-layer light detection and ranging (LiDAR) sensors. The sensors are mounted in a structure so that trucks can pass through without stopping. The proposed system can be used with any type of compact load such as grains, and powders. A mesh of the load is built and used for estimating the volume. A simple, efficient, and effective heuristic is proposed for tracking the truck's positions. The system was deployed and evaluated in a mining company in real conditions of operation. Experimental results indicate that the system produces accurate estimates of the volume of ore powder transported by trucks. The reconstruction of the loads and the estimative of their volumes are performed once the data is acquired and lasts less than 1.5 minutes on average.