Volumetric estimation using drones is a transition from the typical “what” question to the “how much” question of drone vision. Ever since the military made the technical know-how available, people have realised UAVs’ infinite potential. Volumetric and stockpile estimation is a precise and critical activity that drones can complete with accuracy and speed. This is a boon for many industries.

What is Volumetric Estimation?

Simply stated, it means a fair estimation of the volume of the subject. The subject could, in principle, be anything under the sun; however, there is a catch. Estimation is done in cases where accurate measurement is either not feasible or extraordinarily complex and time-consuming.

Largely concerned with the oil and gas and mining industry, volumetric estimation forms the core of all the processes in these sectors, augmenting decision-making and operational effectiveness. Further, the amount of the substance, be it hydrocarbons or mined mineral ores is estimated solely based on their volume.

The mined minerals are bunched up and stored in “stockpiles” of predetermined dimensions for ease of estimation. However, with the existing deterministic and probabilistic estimation techniques, the accuracy is greatly compromised. Also, errors in estimation, both material and human are a cause of concern. To mitigate this, repeated estimations are conducted at different times over a considerable length of time and averaged. This process, apart from not being fool proof is time-consuming and labour intensive.

Mining also calls for regular volumetric estimation of the quarry itself from time to time to monitor progress and regulate workflow. This typically includes elevation profiling by drones, primarily in two ways – photogrammetry techniques and LiDAR (Light Detection and Ranging) technology.

Volumetric estimation is also used as an essential tool in geology and archaeology for functions like terrain analysis, ground water estimation, etc.

A peculiar use case of volumetric estimation where drones can be of immense benefit is disaster relief. The most crucial step in responding to destruction by any natural disaster is to estimate the amount of material needed to rebuild and repair damaged infrastructure. Current practices involve a detailed human survey with unwieldy equipment which may be time-consuming, expensive, and inaccurate. Drones can be employed to obtain stereo imagery which can further be used to estimate volumes in a safe, inexpensive, and accurate manner. Suitably reinforced with terrestrial sensors based on LiDAR technology, 3D imagery of the affected area and also the fill available can be used for accurate volumetric estimation. These characteristics, in conjunction with the remote surveying capability of drones in dangerous areas, could facilitate response to a disaster in less time and with increased accuracy and precision.

Stockpile Volumetrics

Undoubtedly, stockpile monitoring and measurement is an important area of volumetric analysis in mines and thermal power plants. The sheer quantity of minerals stored and transacted daily having high financial stakes makes their fast and accurate measurement essential. However, it is humanly impossible to measure these quantities by weight and hence, the volumetric way is the only feasible way to indirectly estimate the quantity. Also, regular monitoring of the stockpiles is imperative to ensure precise accounting of transactions. At power plants, accurate accounting of strategically stockpiled coal is an essential task, for it is a key ingredient in determining the amount of coal used in the various stages of power generation. The benefit of volumetric stockpile estimation is threefold:

1. The inventory is tracked meticulously for evaluation, financial reporting and excise duty
2. Assessing stockpile volumes determines managing, purchasing and stock administration
3. Improving the stockyard management

Traditional volumetric estimation is a tedious process which involves a detailed manual survey. It involves capturing coordinates, elevation profiling and 3D surface model generation, followed by volumetric calculations/comparisons for each stockpile. Modern approaches to volumetric estimation using drones involves two primary techniques namely, photogrammetry and LiDAR technology. Photogrammetry is a technique where a series of images captured by high-resolution cameras is used to generate 2D and 3D models of objects using techniques of optics and projective geometry. A LiDAR based implementation on the other hand generates a point-cloud (a collection of representation of every point in space with a set of attributes) by accurately computing the position of each point based on the reflections of laser beams transmitted by the onboard sensors. A coloured 3D model is then generated from this point cloud. This technique therefore is more accurate, reliable and efficient while photogrammetry is more cost effective. ideaForge, in a project in collaboration with NTPC was able to establish that LiDAR-equipped drones could reduce time consumption by 60% compared to a photogrammetric drone.

Methodology

Although highly application specific, volumetric estimation methodology generally involves the following components in varying degrees of complexity: –

1. Aerial Survey and Mapping: The foremost aspect of volumetric estimation is aerial survey. Co-ordinated or individual flights of drones provide a series of aerial images which are further processed for mapping. Manual survey of the site is conducted apriori to establish most favourable locations and number of GCPs; and flightpath is determined accordingly. Although photogrammetric drones with high-resolution cameras are effective, use of LiDAR technology greatly increases the accuracy of data as it bases its readings on an active sensor. It actively shoots out laser beams and makes measurements of the time it takes the light to return.
The images/ LiDAR point cloud data obtained from the survey is processed and a map is generated. The accuracy of survey is then verified on the ground and the areas requiring re-survey are identified and surveyed.

2. 3D Model and DEM Generation: The map in conjunction with sensor data is used to generate a 3D Model of the area using powerful tools/ photogrammetry techniques. The software also generates a Digital Elevation Model (DEM) of stockpiles. In case of LiDAR, the point-cloud data is used to generate a colour-mapped 3D model, according to the elevation profile.

3. Volumetric Measurement: A computer-based software empirically computes the volume of the surveyed stockpile using its DEM and mathematical equations.

Benefits

1. Accuracy: Considering the sophistication in terms of technology and ease of deployment, drones can provide highly accurate volumetric measurements than that by traditional methods. Drones when appropriately deployed with the right set of sensors have provided exceptional accuracy upto a few centimeters.

2. Efficiency: Considering the speed of survey and processing in real-time, the drones get measurements done at a much faster rate in a highly efficient way compared to manual labour. For example, in an ideaForge – NTPC collaboration project, drones provided a quick and accurate estimate of stockpiles, which otherwise would have been highly labour intensive and time-consuming.

3. Accountability and theft prevention: Drones offer higher accountability and can prevent illegal mining activities through tighter inventory management and consistent monitoring.

4. Accessibility: Drones can fly over hazardous and hard-to-reach areas and collect imagery without risking the lives of personnel.

As the world moves on to Industry 4.0, these are essential changes that shift the tides of global civilization. Drones for volumetric estimation is a repeatable and scalable application that is the beginning of a long-term domino effect.