A 3.822Mb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE
The LiDAR profession has evolved significantly since LiDAR became a viable mapping solution. In the beginning of commercial LiDAR it was about the technology and what the specific LiDAR technology could do for you. This continued for some time and to a certain extent it still continues. This means that the Mobile LiDAR peeps would say, "you need to use mobile for that" The aerial peeps would say, "you need to use mobile for that" and now the focal plane peeps say, "use us" Now, with UAS available the UAS peeps say you need to do that with UAS. Well, is that really true? Why wouldn’t it be best to provide the solution to a given problem with the best solution possible regardless of technology and this isn’t saying LiDAR is even the right solution but what has happened over the last couple of years is that solution based approaches are much more sellable to end users and we have a great geonerd tool box to provide solutions for our clients. Please note that the geonerd tool box includes all types of remote sensed technology including survey.
The advances of remote sensing technology has made the hybrid approach much more viable and this approach doesn’t require that all the technology has to be co-mounted in the same platform and depending on the solution it may not be an option. It some cases it is more cost effective and more efficient to have separate platforms. Also, the solution requires separate platforms for a variety of reasons. This example maybe stupid but it makes the point. It is not possible to mount a mobile LiDAR and an airborne LiDAR in the same platform but both being part of a Hybrid Remote Sensing approach. This approach is commonly used for Positive Train Control and Autonomous Vehicle solutions. It is important to have an open mind when deciding on what is the best hybrid solution for a problem. It should be noted that when talking about hybrid remote sensing solutions that in actuality Mapping professionals have been co-mounting different types of sensors together for a long time but this become more prevalent with the NERC regulations. We as professionals started thinking about what else could we use these sensors for when NERC began to wind down. Furthermore, it propelled us into thinking of all the options we could utilize. The introduction of new technologies such as Focal plan and UAS furthers the concept of Hybrid Remote Sensing approaches. Additionally, the advances in technology lead to much smaller remote sensing sensors giving the ability to wrap several sensors together when that wasn’t possible in the past.
The following solutions are some of the Hybrid Remote Sensing Technologies currently being used but by no means does this mean there aren’t several others currently being explored. The critical components in determining what should be used in conjunction with LiDAR is based on what is the end user trying to do? What is the horizontal and vertical accuracy requirements? What are the end user requirements? What resolution of data do I need to meet the requirements of the project? What is the end product? Ironically, in most of the examples that will be given the end products being used are much less detailed then the actual data products being generated. This also leads to additional uses for the data for the end user because the detail of the collected data can be used for much more then what its main intent was.
Positive Train Control (PTC)
PTC is a computer system that is installed on a train that makes the operation of the train fully computerize. The system uses GPS on the train to interact with the database generated from the mapping data. Several times a year train derailments make the news on a regular basis. The NTSB believes most derailment is accidental and preventable. Typically, most of the train derailments are a result of operator error. Most all train derailments could be prevented by having a PTC system in place. Currently, PTC is required for all Class 1 railroads. Class 1 railroads are any trains that carry poisonous and toxic hazardous materials and all heavy commuter trains. The federal requirement for compliance was extended to December 31, 2018 as a result of H.R.38.19 Surface Transportation Extension Act of 2015 because the owners and operators requested the extension as a result of non-compliance.
The best way to describe the database is it is an electronic equivalent to a track chart detailing every aspect of the rail system. This includes but is not limited all rail network characteristics, signage, crossings and overhead structure of the network. Currently, depending on the complexity of the rail network, The Hybrid Remote Sensing approach would include HDMS (High Definition Mapping System) LIDAR typically collected at a sample density of 25ppm minimum, Ortho imagery at a very high resolution, Oblique imagery at ridiculous resolution, Georeferenced video, survey and either one of the following, HDS LiDAR (terrestrial), Mobile LiDAR, or UAS information. The last being determined by a given project and cost analysis. Typically, the survey information density is significantly at a higher requirement then typical mapping and corridor projects.
The vertical and horizontal requirements for PTC are very lose compared to typical mapping or surveying projects and the Hybrid LiDAR approach far exceeds the geometric accuracy requirements based on the published specifications for this type of work and the reason for that comes for the required precision for these projects. We need the precision so that we can see all the assets with in the rail network. This mapping requires that you can see all text on all signs, switches and rail detail. So basically, with this approach you can see the spikes that hold the rails on the ties. Typically, the precision of this data sets are down to a 1cm or less and the accuracy as a result is in the neighborhood or 2 to 3 centimeters and the required accuracy for PTC is 7 feet horizontally and 3 feet vertically.
Autonomous Vehicle Mapping (AVM)
AVM is self-driving vehicles including all types for CARS, Trucks and Vans for those who are not familiar with it and this is not meant to be condescending, just making sure. All automakers and others using or planning to use selfdriving autonomous vehicles are involved in this application such as GOOGLE, APPLE and UBER. It is fair to say that fully autonomous vehicles will not be here for some time based on insurance and regulatory issues. It is also fair to say that driver assisted self-driving vehicles are here and that most new vehicles have some sort of autonomous functions in a varying degree currently. The same technology we are using in the mapping profession is currently being utilized in the autonomous vehicles but these sensors are a much cheaper version of the technology. There is some indication that currently and in the near future the sensors used in the vehicles are providing information to make the vehicles smarter and provide learned data to help make the vehicles drive better in given situations. This is similar to what IBM’s Watson does as a super computer.
The most notable mapping technology for AVM is mobile LiDAR and currently several automakers have spent several million dollars on creating databases for this data for Autonomous vehicles. Like PTC the original data gathered for this application is much more then is needed for the actual application and all that information could not be reasonably stored on a car. Additionally, many other data sources are used for AVM such as information that is available and used in a smart phone navigation system. It is critical for the Vehicle to know speed limits and traffic information that can be extracted from a navigation system.
Currently, this application and the solutions associated with it are still developing and the most cost effective hybrid remote sensing combination is still being developed. The question that should be asked is, as this technology and remote sensing technology continues to develop is when is information or data too much data and do we really as end users need all that information because there has been suggestions that a similar approach that is used in PTC could be applied to AVM. They are very similar solutions but the requirements for AVM is more strict in regards to the accuracy. The main two differences is that all the information for autonomous vehicles needs to be relative to the vehicle itself and the vehicle needs to be able to evaluate and react to a significantly changing environment around it such as other vehicles, new development, and pedestrians.
There are many more hybrid remote sensing applications such as highway safety and airport mapping but the important component is to realize that a given technology isn’t the best solution for an application just because a given provider with a given technology says so. The mapping profession provides a very large geonerd tool box and you don’t use a screw driver to cut a board.
James Wilder Young (Jamie) CP, CMS-L, GISP is currently a Senior Geomatics Technologist for Merrick & Co. located in Greenwood Village, Colorado. He is currently supporting all aspects of LiDAR technology development and is lead for all HDS and Drone functions. His experience includes all aspects of LiDAR including sensor development, applications development, data acquisition, data processing and project management. He graduated from The University of Colorado.
A 3.822Mb PDF of this article as it appeared in the magazine complete with images is available by clicking HERE