Ask Your Mates Open Forum

LIDAR for roadway asset management

  • 1.  LIDAR for roadway asset management

    Posted 28 August 2013 16:45
    Hi All,

    I'm wondering whether anyone is using LIDAR as a regular data source for asset inspections, or to collect road-related inventory?

    My Director was recently taken on a ride-along with a mobile LIDAR vendor and is interested in finding out whether this technology could be used on an ongoing basis to support/replace existing condition surveys (pavement, sidewalk, kerb & channel, streetlights, signals, fences, etc), and to update/validate our asset registry.  I'm skeptical about the quality/accuracy/precision of data available, and suspect it wouldn't be a suitable replacement for any of our existing functions or processes.  As such, I'm loathe to spend too much time investigating the technology unless someone has a real success story to report (vendor websites claim it's great for asset management, but is it really?).

    As a desktop location reference, Google Streetview (free!) works pretty well, so not sure we need more than that, particularly given the likely initial and ongoing cost and training requirements (data collection once or twice a year, data storage, new software, new analysis skills, etc) associated with introducing LIDAR.

    I could see a highway authority using LIDAR to avoid having to drive for hours (or days) to inspect remote locations, but in Calgary we can get to any location within half an hour.

    Anyway, if anyone has tried this out, I'd love to hear what you used it for and whether it was a good experience (or not).

    Thanks all, and cheers,

    -------------------------------------------
    Brian Arthur
    Asset Mgmt Coordinator
    City Of Calgary
    CALGARY AB

    -------------------------------------------
    BlogPageSpacerBlank


  • 2.  RE:LIDAR for roadway asset management

    Posted 31 August 2013 07:02
    I will be very interested in any response as I would like to know more about the mobile application

    There are a lot of inspections we are required to do that are not actually done well due to the difficulty in undertaking them - I think there is potential for this application if it is cost effective.

    -------------------------------------------
    Neil Bennett
    Road Asset Manager
    Fulton Hogan Ltd
    Christchurch NZ

    -------------------------------------------






    BlogPageSpacerBlank


  • 3.  RE:LIDAR for roadway asset management

    Posted 31 August 2013 07:07
    Hi Brian,

    ARRB Group has been asked by the Australian state road agencies to assess the value of mobile LiDAR for asset management. Our experience is that accuracies of the available LiDAR system vary, but for the applications you mention, the required accuracies can be realised (with the exception pavement quality measures like IRI). This means that your concerns about quality/accuracy/precision of data available are a matter of choosing the right system for the job. If you would like to consider using mobile LiDAR to replace existing condition surveys, the next questions are:

    • which mobile LiDAR systems are suitable for surveying sidewalk, kerb & channel, streetlights, signals, fences, etc?
    • what are the savings of replacing (some of) your current surveying activities?
    • what are the benefits of additional uses of mobile LiDAR data?

    The experience of US and Australian road agencies with mobile LiDAR has been discussed in a workshop with industry partners.

    Key messages were:

    Potential benefits of MLS is that a single acquired dataset can be used for a variety of applications
    Typical examples of many possible applications are:

    1. (Bridge) clearances
    2. Line markings/lane widths
    3. Sign extraction using LiDAR and video
    4. Asset inventory
    5. Pre-design

    - There are different types of mobile LiDAR systems, for example varying in accuracy. Different types have different uses.
    - Also, mobile LiDAR is just one tool, which needs to be augmented by stationary/aerial etc
    - Costs vary largely and depend on:

    1. Intended use (number and type of features to be extracted)
    2. Required accuracy and density
    3. Required level of processing

    Main issues for road agencies to start using mobile LiDAR:

    1. Limitations to automated data extraction
    2. Trust that the proposed solution is fit-for-purpose
    3. Commitment from different departments for multiple use
    Hope that helps!

    Cheers
    -------------------------------------------
    Kim Sedgwick
    Operations Manager (Systems)
    ARRB Group
    Vermont South VIC

    -------------------------------------------






    BlogPageSpacerBlank


  • 4.  RE:LIDAR for roadway asset management

    Posted 31 August 2013 07:13

    We've been doing a number of road surveys using mobile laser scanners (slightly different to LiDAR I believe). The most recent one I've been involved in is scanning 60km of road before the construction of the Ichthys Project. The NT Gov were concerned that the high number of road train movements may damage the road, so we've done the scanning so we have an accurate "before" record, and can compare it later on or at the end of the project to an "after" scan. The can pick everything up in a great level of detail, and it takes 4 high definition photographs every 10m, effectively forming your own streetview - which is has come in handy for seeing if any signs etc are damaged. There's a lot of amazing stuff that can be done with the data, and can be presented in a number of ways including using free software (Leica TruView) that allows you to take measurements etc

    I'd contact Leigh Finlay for more information (and a lot more examples of the work that we've done for various road authorities) - LFinlay@globalskm.com +61 (0)2 9928 2540, Leigh is our "Practice Leader" for survey work. You can then see if something like that will suit your needs or not.

    Hope that helps,
    J-P
    -------------------------------------------
    John-Paul Foster
    Senior Civil Engineer
    Sinclair Knight Merz - Darwin, NT
    Parap NT au

    -------------------------------------------






    BlogPageSpacerBlank


  • 5.  RE:LIDAR for roadway asset management

    Posted 31 August 2013 07:13
    Why would you use lidar (difficult to visualise) when you can use GPS referenced video?

    -------------------------------------------
    George JasonSmith
    AECOM NZ Ltd
    Christchurch

    -------------------------------------------






    BlogPageSpacerBlank


  • 6.  RE:LIDAR for roadway asset management

    Posted 03 September 2013 17:47
    Yep, GPS referenced video is a great technology, very useful - however there are some limitations that LiDAR can help overcome. For instance, it's very difficult to accurately measure the clearance under a bridge - or say the height of a street sign, or power pole from video alone. ------------------------------------------- Aaron Steinmetz ARRB Group Vermont South VIC -------------------------------------------
    BlogPageSpacerBlank


  • 7.  RE:LIDAR for roadway asset management

    Posted 05 September 2013 18:46

    Brian,

    Contours derived by LiDAR is useful but do not define breaklines well, such as the top and bottom of kerbs, walls etc.

    It is a great tool for area analysis, but has limitations from a brownfield design perspective. The precision is very dependant on GPS acquisition and weather (wind conditions). Stirling uses LiDAR to contours in the City's GIS (removing buildings from the dataset), but survey field tests indicate that 0.1 to 0.2m  vertical accuracies are achievable. LiDAR provides @ 4.5 laser points per square meter, however the points are scattered so that without good digitising of breaklines accuracies may not be improved.

    LandGate in WA are spruiking EarthMine which combines GPS and Video and the ability to take sizes and height and condition from the video. As the program uses triangulation from a short base-line of the vehicle, positional accuracy degrades with distance from the vehicle. However in the immediate road reserve this should not presenta problem except for groups assets such as signs.

    I'm concerned that a 20mm tripping hazard on a path at distance could not be identified. But there is some merit in a yearly video for sign condtion and gross road defects in the road reserve  (So either an officer in the field or the office) . Obviously image resloutiuon is the last factor that needs to be considered,. The data that needs to be held on a server is huge (Terabytes) and will beed to be manged.  

    An estimate for a Council the Size of Stirling   @100 KM2 is around $100 000 P/A, for Earth-Mine so the sums have to add up.

    Regards,



    -------------------------------------------
    Greg Duggan
    Principal Engineering Surveyor
    City of Stirling
    OSBORNE PARK WAau

    -------------------------------------------






    BlogPageSpacerBlank


  • 8.  RE:LIDAR for roadway asset management

    Posted 10 September 2013 16:36

    Thanks for the replies folks, your comments and time are much appreciated.

    What I haven't heard is a resounding 'Yes, we're using mobile LIDAR data on a regular basis to support activities x, y, and z, and it's fantastic' (please correct me if that's wrong).  As such, I'm guessing that it hasn't quite developed to the point of being a key part of everyday roading business, but that it is getting closer.

    I've gathered here in Calgary that the costs of LIDAR data collection have dropped dramatically over the past 5 or 10 years (by a factor of 10 by one estimate), and that the technology is becoming far more prevalent (I was told that there are now 3 LIDAR vendors situated in Calgary, where 5 years ago there were none).  However, because of the high costs of data storage, we would likely need to consider non-standard approaches, such as having a consultant store the data, or putting it on a few hard drives, and those solutions bring their own issues.

    To give some context to Calgary - we have just under 5,300 centreline kms of roadway, of which 1,100 are expressway or arterial (overall we have about 15,100 lane kms).  So it's rather a large network even for just the bigger roads, and there would be a LOT of data to collect and store if we were to capture the whole network.

    Re: the accuracies that we are currently working with, I've heard that they *might* be achieved if we select the right system, but it sounds like we would be looking at the higher-end systems in order to achieve that.  I guess a key judgement we would need to make is 'how close is close enough' - which would need a full test run in order to document and verify accuracies and issues against real conditions.

    All in all, a shiny toy that's very neat, has some immediately useful applications and has loads of potential, but probably not quite there yet in terms of fully supporting our needs.

    The City of Calgary is now collecting airborne LIDAR/imagery, but in Roads we haven't explored the data analysis potential of that yet, we just use the resulting imagery as an enhanced orthophoto behind our GIS layers (the imagery can be viewed from an oblique angle, which can be quite helpful).

    If we do pursue a full investigation of mobile LIDAR, I'll post at least a summary of the results.

    Thanks again!

    Cheers,
    Brian


    -------------------------------------------
    Brian Arthur
    Asset Mgmt Coordinator
    City Of Calgary
    CALGARY AB

    -------------------------------------------






    BlogPageSpacerBlank


  • 9.  RE:LIDAR for roadway asset management

    Posted 08 September 2013 23:13
    A LIDAR survey (3D scanning) from a moving vehicle is not at this moment capable of picking up a 20mm trip hazard on a footpath.
    Lidar accuracy is +- 2cm for most systems, Stationary LIDAR is sub 2cm distance dependent.

    LIDAR is not yet any way accurate enough to do roughness or rutting on road pavement surfaces even from a stationary scan,
    Other issues to deal with are similar to vehicle based video collection such as parked cars etc blocking the LIDAR.

    As Greg points out data sets are also rather large at the moment.
    There are moves to auto extract kerb & channel etc from LIDAR but again see the above issues to start with.

    Regards

    -------------------------------------------
    Graeme Chalmers
    Team Leader Data Collection QLD/NT
    ARRB Group

    -------------------------------------------






    BlogPageSpacerBlank