Blog | Advancements in UAV Hardware and Photogrammetry Software Applications, Combined with the Power of Virtual CRASH, Provide Substantial Savings to the Public Safety Sector
Looking through quotes for 3D scanner-based solutions for public safety can certainly cause some amount of “sticker shock.” Many times, these quotes run in the hundreds of thousands of dollars! Below is a typical quote for law enforcement from a laser hardware manufacturer.
Notice, the significant cost for hardware. Also notice the separate costs for the point registration software and the diagramming software and of course the 3-year cost for both tech support and “maintenance” on both of these. It all quickly adds up to huge sums of money which will break the budget!
Though the millimeter level accuracy of a typical 3D laser scanner is naturally appealing to the forensic investigator, the measurement accuracy gained per dollar spent may not make sense for the budgets of all law enforcement agencies. How can this be? Consider, the typical driver of uncertainty in a conservation of linear momentum calculation isn’t the ability to resolve post-impact trajectory lengths to the nearest micron, but rather, it is typically dominated by issues such as the appropriate tire-road drag factor or the vehicle inertial properties to use in the calculation. To bring the point home, just try this exercise: first, calculate the pre-braking speed needed to stop a vehicle over a distance D= 50 ft assuming a drag factor of f = 0.7. Pretty simple. You should arrive at a pre-braking speed of about 32.4 mph.
$$S \approx \sqrt{30 \times D \times f} $$
Next, repeat this calculation, but use D = 53.133 ft. You should arrive at a pre-braking speed of about 33.4 mph.
These results tell you the following: in order to nail down the slide-to-stop speed to better than 1 mph uncertainty (when braking over about 50 feet is involved), the error in your distance measurement (D) from scene data better be smaller than 3 feet! Using a scanner in order to ensure that you can resolve distances better than 3 feet might be overkill here. Similarly, if your department uses energy methods to arrive at closing speeds (CRASH3), and you want to use point clouds of the subject vehicles to help estimate vehicle crush profiles, your calculations will be completely dominated by the uncertainty in stiffness coefficients (A and B) which are usually known to only 25% to 50%, and not the crush measurements themselves. We know plenty of recons that still use tape measures for their crush profile measurements!
Remember, laser hardware manufacturers are typically publicly traded, and as such, their behavior is predictably driven by the desire to maximize quarterly sales figures; thus, the formidable sales forces of these companies will tend to steer discussions away from these practical considerations of the needed resolving power for reasonably accurate accident reconstruction calculations (if the sales force is even capable of having such discussions), and toward trying to instill into their sales prospects a sense of “F.U.D.” (fear, uncertainty, and doubt), by focusing discussions on supposed measurement inaccuracies from which their products will presumably help save you.
UAVs
We have now entered the era of the drone. Studies show, using a drone with a photogrammetry-based software solution yields results that are more than suitable for accident reconstruction analyses, especially when ground control points (“GCPs”) are captured by total station or RTK GPS, where the uncertainty on distances is typically on the order of an inch or less [1,2,3,4]. For the crash investigator, the speed with which a drone mission can be completed, and normal traffic flow can be restored, is hard to beat - 20 to 30 minutes at the scene in typical cases. The cost of a new drone is also hard to beat. Keep in mind, you’ll also need to purchase a photogrammetry application, of course, which will process those hundreds of drone photographs in order to produce a spectacular bird's-eye view “orthomosaic” aerial image as well as the 3-dimensional point cloud representation of the scene.
Here we see a typical proposal for a drone-based hardware solution, including a photogrammetry application and, of course, Virtual CRASH for diagramming, animating, and simulation. See our hardware page for pricing on popular drones.
Notice, this example quote is one-tenth the cost of above for a scanner! For lower tier scanner packages, the potential savings may drop to about one-fifth (rather than one-tenth) the cost by going to a drone-based solution. Even with the added cost of a total station or RTK GPS receiver with data collector, the total cost to for an all-inclusive drone, photogrammetry, GCP collection device, and Virtual CRASH is vastly lower. This quote includes the powerful Virtual CRASH software application, which includes the ability to diagram, animate, and simulate with point cloud data, with no additional fees for point cloud importing or for “maintenance”. Virtual CRASH software updates are free. Factoring in the cost of ownership over a 3 to 5 year period, the one-time purchase of Virtual CRASH, with the low upfront costs of a drone and a photogrammetry application, makes this type of package an extremely appealing alternative to scanner-based solutions. To learn more about using Virtual CRASH with a drone and photogrammetry based solution contact us.
References
[2] Comparing the Accuracy of Image Based Scanning Techniques to Laser Scanners, C. Grimes et al, SAE 2018-01-0525.
[3] Do more GCPs equal more accurate drone maps? Summary of study conducted by the Nevada DOT.
[4] Small Unmanned Aircraft Systems Photogrammetry vs Total Station, Sergeant Joseph Weadon, Missouri State Highway Patrol, Collision Vol. 13 Issue 1.
