Application of Aerial LiDAR Technology in Indian Railways – a Case Study on Mumbai-Ahmedabad HSR Corridor Project

Dr. Bharat Lohani, Professor (on leave), IIT-Kanpur, and Director, GeoKno India Pvt Ltd (a GMR Group Co.)

Dr. Bharat Lohani is Professor (currently on leave) from IIT-Kanpur. His research involves use of high resolution remotely sensed data, especially LiDAR data, for 3D modelling, visualization and various other applications. He is also using GIS for innovative applications and in the following specific areas:- 3D Laser imaging and LCS measurement, Methodology development for motion correction and error analysis in laser scanning, Propagation modelling using high resolution LiDAR data (flood, sound, GPS signal), Mapping of buildings, specifically urban and those of historical significance, using terrestrial laser, Laser scanning of complex structures, Development of LiDAR simulator-airborne and space-borne, GIS for optimized land consolidation in agriculture, Landmark based advanced navigation under GIS environment.

Dr. Lohani has Professional Affiliations at Institution of Surveyors, India, Indian Society for Remote Sensing, Institution of Engineers, India. Dr. Lohani has won various prestigious awards at the national and international levels, such as:

  • ISRS National Award for Geospatial Excellence 2012, Presented on 4th December 2013 during ISG-ISRS Annual Symposium at Vishakhapatnam
  • Best paper award at INCA Congress 2012, Dehradun for paper “Flash LiDAR based surveillance system”.
  • Best ISBA Start up in ICT category in 2011 for Geokno India Pvt Ltd, being incubated at IIT Kanpur and founded and mentored by me.
  • Silver award for Limulator2.0 for standing second in CATCON-5 during XXI ISPRS Congress, Beijing 3-11 July 2008.
  • Best Poster Award at ISPRS conference on Optical 3D measurement techniques, 9 – 13 July 2007, ETH Zurich, Switzerland.

Dr. Lohani has several publications to his credit viz.

  • Goel Salil and Lohani Bharat (2014), A Motion Correction Technique for Laser Scanning of Moving Objects, IEEE Geoscience and Remote Sensing Letters, Vol:11, PP:225-228
  • Ghosh S and Lohani B (2013), Mining LiDAR data with spatial clustering algorithms, International Journal of Remote Sensing, Vol 34-14, PP: 5119-5135.
  • Dashora A, Lohani B (2013), Turning mechanisms for airborne LiDAR and photographic data acquisition, Journal of Applied Remote Sensing, Vol: 7(1), 073488.
  • Dashora A, Lohani B and Deb K ( 2013), Two-Step Procedure of Optimization for Flight Planning Problem for Airborne LiDAR Data Acquisition, International Journal of Mathematical Modelling and Numerical Optimisation.
  • Yadav M, Goel S, Singh A K, and Lohani B (2013), Mobile mapping system using laser scanner, GPS and IMU, ISRS.

Application of Aerial LiDAR Technology in Indian Railways – a Case Study on 509 Km Mumbai-Ahmedabad HSR Corridor Project

NEW DELHI: Engineering Survey is among the most important activities during pre-feasibility and Detailed Project Report (DPR) stage, as ground levels and terrain profile are fundamental inputs for accurate Earthwork calculation and design of the project components. Hence, fast, accurate and comprehensive topographic data are the primary requirement for successful project execution. Absence of good quality topographic data or delay in obtaining these data results in problems at various stages of project, for example, procurement/contract, project planning, project execution, operation and maintenance.

This article introduces Airborne LiDAR Technology and its use in Railways, which has now become standard around the world. LiDAR gives direct measurement in the form of (x, y, z) co-ordinates of innumerable points on the ground thus capturing the geometry of all objects and terrain precisely. LiDAR data are being used in multiple applications which require topographic information of the terrain. Considering comprehensive capture of ground and objects by LiDAR, the data captured can be used in multiple applications related to Railways thus eliminating the need of multiple different surveys in field.  Besides DPR preparation, LiDAR data are also being used for the development of GIS based asset management system for Railways.

Project Background

The 509 Km Mumbai – Ahmedabad High Speed Rail corridor is the India’s first high speed rail corridor which connects the financial capital & Ahmedabad. So for India doesn’t have any railways classified as High Speed Rail, which has operational speeds in excess of 200 km/h. Now India has experienced the Semi-High Speed Trains “Vande Bharat Express,” flagged off for an inaugural run by Prime Minister of India Narendra Modi, on 15 February 2019, with its commercial run started from 17 February 2019 onwards. The train has Maximum speed of 180 km/h (110 mph) and replaced Shatabdi Express and Gatimaan expresses which connects Delhi & Agra traveling at a speed of 160 km/h.

In order to achieve better commuter experience in the country, the Prime Minister has envisaged for execution of High Speed Rail lines and eventually incorporated National High Speed Rail Corporation Limited (NHSRCL) as a subsidiary of Rail Vikas Nigam limited. India and Japan International Cooperation Agency (JICA), a joint venture consisting of Japan International Consultants for Transportation Corporation Ltd and NIPPON KOEI Corporation Ltd, signed the Memorandum of Understanding (MoU) to implement & technically assist the Mumbai – Ahmedabad HSR. NHSRCL has associated with RITES to complete the feasibility study to propose the alignment & carry out the planning for the establishment of HSRC. A joint feasibility study has been conducted by JICA & RITES. Final alignment has been proposed to Railway Ministry according to the feasible study report.

RITES called for an open tender to conduct the Aerial LiDAR Survey along the center line of proposed alignment of Mumbai – Ahmedabad HSRC on QCBS two packet system. Across India, Top GeoSpatial companies were participated in the open tender process. Among them, Geokno India Private Limited, a GMR Group Company has satisfied all the qualification criteria’s as mentioned by the RFP issued by the RITES Limited. Upon evaluation, M/S Geokno India Private Limited has been awarded to carry out the survey along the center line of proposed alignment of Mumbai – Ahmedabad HSRC. Aerial LiDAR Survey complemented with Aerial Digital Photography has been carried out for the proposed alignment of center line. Ground Control network was established with reference to the existing control points provided by RITES. For HSR alignment, 150 m wide corridor on both sides of the proposed center line has been covered with Aerial LiDAR. For River / Stream crossings, 1 km wide corridor perpendicular to the proposed center line on both sides has been covered with Echo Sounder, Total Station & DGPS.

Project Area

The Mumbai-Ahmedabad high-speed rail corridor is a high-speed rail corridor project of connecting the cities of Mumbai, Maharashtra and Ahmedabad, Gujarat in India. It will be India’s first high-speed rail line. The alignment of the HSRL will pass through 3 districts (Mumbai, Thane and Palghar) in Maharashtra and 8 districts (Valsad, Navsari, Surat, Bharuch, Vadodara, Anand, Kheda and Ahmedabad) in Gujarat. The proposed alignment also passes through a small area section falling in the UT of Dadra. Schematic map of alignment along with details have been provided below:

Working Principle of LiDAR Technology

A LiDAR sensor measures the range from an aircraft to a point on the ground by using the time of flight of a laser pulse transmitted. The range measured is converted into the co-ordinates of the point hit by laser pulse using on-board GNSS receiver and IMU sensor (Figure 1). As LiDAR sensor scans the ground underneath, the coordinates of millions of points on the ground are generated every second.

Continuous operation of a GNSS Base Station on the ground is essential during LiDAR sensing for supplying corrections to the onboard GNSS, thus determining the position of onboard GNSS accurately. A digital camera is also flown with LiDAR sensor to capture high resolution images of the terrain. The net result is a large number (tens of billions) of points with their known co-ordinates along with colored images of the area flown. Unlike other technologies, LiDAR provides the coordinates directly and is also able to see under forest canopies.

In general, an Airborne LiDAR can collect over 5,00,000 pulses in a second with 5-15 cm absolute accuracy in vertical and 10 to 20 cm absolute accuracy in horizontal. LiDAR captures coordinates of everything present on the ground, thus creating a 3D replica (like photocopy) of the terrain. The coordinates are generated for all objects, viz., Buildings, Roads, High Voltage transmission lines, Bunds, Bridges, Culverts, Canals, Railway Tracks, Trees, Bushes, Hedges, Water bodies, Railway Stations and Yards, Poles, Overhead transmission lines, Boundary walls etc.

Advantages of LiDAR Technology

Traditionally, Railways has been employing conventional surveying technologies, like Electronic Total Station (ETS) Survey, GNSS Equipment Survey (Static, Kinematic (RTK, PPK)), Satellite Imagery based survey and recently Unmanned Aerial Vehicle (UAV) or Drone Survey. Therefore, it is appropriate to compare these technologies with the LiDAR technology for assessing the advantages of the latter. Table 1 clearly demonstrates the suitability of airborne LiDAR technology over other methods when comparison is done with respect to several important parameters, listed on first column. The green color indicates advantageous position of technology while red indicates opposite.

A very significant advantage of LiDAR, unlike other technologies, is that LiDAR travels through the gaps in tree canopy and measures the coordinates of the ground under tree canopy. In addition, LiDAR provides the coordinates of several points on tree canopy thus leading to the determination of the height and biomass volume of tree.

For planning and design for railway lines which pass through heavy forest, like in North-East States, Himalayan Region, Jammu & Kashmir, Chhattisgarh, Jharkhand, Madhya Pradesh, Western and Eastern Ghats etc. where other methods of survey fail, LiDAR can provide very good solution. Interestingly, LiDAR also maps objects as thin and inaccessible as transmission lines thus generating necessary information for design work of new electrical installations or overhead transmission lines during widening of existing railway lines.

LiDAR Data Products and Their Use for Railways

The objects identified on LiDAR point clouds generally can be classified into ground, vegetation (high, medium, low), buildings, water, transmission lines.  Different derivatives can be generated which are used in Final Location Survey (FLS) and Detailed Project Report (DPR) purposes.

Digital Surface Model (DSM), defined as the 3D model of the ground including all objects like buildings, trees, towers etc., can be easily generated using first-return LiDAR data. This is useful for Railways for analysis of alignment and identifying the obstructions, e.g. houses, trees.

Digital Elevation Model(DEM) is generated by using only those LiDAR coordinates which are on the ground. LiDAR is the only technology which can provide accurate DEM. All other features other than ground are removed from the classified pointcloud data thus resulting in accurate Digital Elevation Model (DEM). Contours are generated up to 30 cm Contour Interval (CI) or better using the classified ground points.

LiDAR DEM and simultaneously captured images are used to produce accurate Orthophotographs at 5 cm to 15 cm resolution, as per need. An orthophoto is like a map in geometry while a photo in visual appearance. This is highly useful to interpret ground details and carry out measurement without going to ground.

Various features, e.g., Buildings, Roads, High Voltage transmission lines, Bunds, Bridges, Culverts, Canals, Railway Tracks, Railway Marshalling Yards, Trees, Bushes, Hedges, Water bodies, Railway Stations and Platforms, Poles, Overhead Transmission lines, Railway Traction & OHE, Boundary walls etc., as required for a project can be extracted using LiDAR point cloud and orthophotograph to generate maps up to the scales of 1:2500. These features then can be represented in any GIS or CAD format for their further use in project planning and execution.

Case Study: Aerial LiDAR Survey for 509 Km Mumbai-Ahmedabad High Speed Rail Corridor

High Speed Rail Corridor Limited (HSRCL) through its consultant RITES chose LiDAR technology for mapping entire corridor. Their selection was based on the need of accurate, fast and comprehensive data. One additional reason for choosing aerial survey was that it does not affect the local populace and the work can be completed unhindered. The availability of complete 3D replica of the ground in the office computer, through LiDAR survey, is also advantageous to project designers as they can focus on design instead of wasting time in repeated field visits for verification, as happens if survey data are not accurate and complete. HSRCL entrusted the responsibility of LiDAR survey to Geokno India Pvt. Ltd. which is a spin off from IIT Kanpur and is now a subsidiary of GMR group.

Instrument Employed

Geokno employed Lite Mapper 7800 system which consists of Airborne Laser Scanner RIEGL- LMS Q 780, Phase-1 100 MP Digital aerial camera, IMU and GNSS Systems. This system makes use of a laser source, Multiple-Time-Around (MTA) processing, echo digitization and waveform analysis. This combination allows the operation at varying flight altitude and is ideally suited for aerial survey of complex terrain. The rotary wing aircraft Eurocopter was employed for flying these sensors over the project area. Dual frequency (L1 and L2)-multi-channel GNSS equipment were used for generating data at 1-second epochs.  The aerial platform and LiDAR sensors deployed are shown in Figure 8 and Figure 9.

Field Work

Initially, a reconnaissance survey was carried out to understand the nature of the project area which helped in further planning and designing of the project. Necessary clearances/permissions for flying over the AOI were obtained from the Director General of Civil Aviation (DGCA), Ministry of Defense (MoD), Government of India, and other concerned agencies to undertake Aerial LiDAR Survey. This was followed by flight planning which ensured that the designated flight paths cover entire study area including enough cross flight lines to eliminate shadowing and allow for proper quality control. Proper Flight line overlaps were planned to ensure that there are no data gaps between the usable portions of the swaths.

Data were captured with reference to GNSS Ground control network established along the HSR corridor. RITES provided Ground Control Points (GCP) of the existing network (300 points) which are situated about 1.5 km to 3 km from center line of the alignment. GNSS Survey was used for establishing additional GCPs to achieve the desired accuracy. Levelling survey using Double Tertiary (DT) method was carried out for about 1200 km thus providing height from SOI vertical datum to all GCPs.

LiDAR Technology for surveying the new Railway lines which is one of the most advanced technologies used by Railways

For LiDAR data pre-processing and quality control purposes 81 Master Control Points (MCPs), 233 Secondary Control Points (SCPs) and 103 of LiDAR Target Points (TP) were established. TPs were established on flat surfaces and marked by paint so the same were clearly visible from aerial LiDAR platform. The horizontal accuracy GNSS work was of the order of 1:100,000. To provide corrections to onboard GNSS during data collection, base stations consisting of two GNSS Equipment were established, providing redundancy and flexibility of capture. These base stations were so established that the aerial platform was always within 25 km from these. The project extent required establishment of 13 such base stations. LiDAR data and photographs were captured for 100 m on both sides of alignment.  This width should be more for complex terrain where chances of shifting of corridor in later stages is higher.

Data Processing

The first return points and the points on the ground, on interpolation, resulted in DSM and DEM, respectively. The images captured were processed along with DEM to result in ground level Orthophoto. Features were extracted from LiDAR Point cloud while taking visual information from Orthophotographs and classified as per the ASPRS Standards and RFP Specifications into Ground, Buildings, Road, Vegetation (High, Medium and Low), Transmission Lines, Water Bodies. The extracted features were layerized with scheme and symbology provided by the client. Topographic Maps were produced at 1: 2500 scale. Profile Sections and Cross sections were prepared as per client approved layout for 283 locations using LiDAR data captured.

Quality Control of LiDAR Data

LiDAR points were compared with independently surveyed GCPs. Methods as per international standards were employed for this purpose. The LiDAR data were found to have an RMSE of 4 cm (9 cm at 95% Confidence Level) in Elevation against the RFP specification of 10 cm. Further, the horizontal accuracy of LiDAR data was 6.45 cm as RMSE (i.e. 11cm at 95% Confidence Level) against the RFP specification of 15 cm. In addition, the DEM generated was also subject to accuracy analysis by comparing the height of corresponding points in ground and was found within specifications.

Deliverables Produced for Use in High Speed Rail Corridor (HSRC) Project

The final project deliverables for HSR project included the following:

  • Raw and Classified Point Clouds with less than 10 cm and 15 cm vertical and horizontal accuracies
  • DEM and DSM at 0.5 m grid size with 10 cm vertical accuracy in GeoTIFF file format
  • RGB and Intensity Image at 10 cm GSD
  • Orthophotos at 10 cm GSD
  • Three-dimensional topographic survey drawing including contours (0.5 m Contour Interval) at a scale of 1:2500
  • Longitudinal sections at 500 m and Cross Sections at 20 m intervals.

Benefit of LiDAR Technology for Ahmedabad-Mumbai High Speed Rail Corridor Project

The LiDAR project is now completed, and all data have been handed over to RITES and HSRCL, who are using these data in their design and planning. RITES and HSRCL have found LiDAR data highly useful for HSR project. A few important use cases are:

  1. LiDAR survey was done without attracting the attention of local populace thus avoiding any conflict with locals.
  2. LiDAR survey has provided 3D digital twin of the entire corridor. This is highly useful for design and planning team as in case of any confusion regarding terrain one can easily verify it using LiDAR and image data in the office thus avoiding repeated visits to field, as in case of conventional surveys.
  3. As only single instrument and party was utilized for entire project there is uniform data quality and accuracy across entire project.
  4. Data for complex and inaccessible areas are also available with the same precision and reliability as in simpler areas.
  5. One-time LiDAR survey has been able to eliminate the need of multiple surveys for this project, for example tree count survey, building plan survey, land plan survey, contour survey, as all these can be mapped correctly and with high reliability on LiDAR data.

Conclusion

Indian Railways is faced with a special challenges now – more and more new projects are coming in very difficult terrain and/or require extremely good quality of data and most importantly have to be completed fast.  Relying on traditional or non-appropriate technologies for FLS survey would lead to inaccuracies or delays.  Airborne LiDAR survey along with digital images provide a solution to cater to this need.  The use of LiDAR technology has been demonstrated around the world and also in India in railway projects.  LiDAR survey not only produces data for FLS purpose but also proves very useful in several other areas like environmental clearance.  Considering the advantages of LiDAR over traditional survey methods and the need of fast and accurate project completion the future will witness more application of LiDAR survey in Railways.

Facebooktwittergoogle_plusredditpinterestlinkedinmail