l HD Mapping Process l

◼︎

Planning



Define target area


An area for geofencing is selected and surveyed to determine optimum functionality at deployment sites.


Potential sites are evaluated based on long term strategic goals with a focus on partnerships and scalable deployments.

 

UMO engages in a comprehensive and collaborative planning process and involves actively partnering with private and public sector bodies to guarantee deliverables and successful deployments.


UMO aims to facilitate interactions between, creators, corporations and institutions to align interests multilaterally to the benefit of all stakeholders.

 

 

 

 

◼︎

Acquiring GCP


Generate GCP (Ground Control Points) in geo-fenced area and perform precise surveying

Ground Control Points (GCPs) are defined as points on the surface of the earth of known location used to geo-reference imagery.


Surveying Pyeonghwa-ro in Jeju-Do,

South Korea


 

 

Surveying Pangyo, Seongnam,

South Korea


 

 

Surveying Yeongdong

Expressway


 

 

GCP’s are the anchor points in which a geo-fenced area is precisely defined and referenced


◼︎

Scanning



Operate mapping vehicles with mobile mapping system installed


UMO utilizes fleet based LiDAR sensors to collect a broad variety of data on the built environment, including:

•Roads: Lane position and regulation, Road boundaries, stop lines, centerlines etc.

•Infrastructure: median strips, tunnels, bridges, underpasses, traffic signs, traffic lights etc.


These sensors operate continuously to provide precise measurements foundational for effective HD maps.

UMO’s mobile scanning process simultaneously collects 3D spatial and locational navigation critical data:

•Centimeter accurate 3D models

•Panoramic street imagery


The following raw data is formulated into highly versatile base maps, that when coupled with UMO’s change-management solution, allows for a wide range of future mobility applications within geofenced areas

 

 

 

 

 

 

 

 

 

◼︎

Scanning



Collect raw data of roads with mobile mapping system



UMO’s fleet based scanning process contextualizes collected road data for the purposes of autonomous localization, perception and planning via:



•Real-time GPS tracking

•Vehicle safety analytics

•Trip video archive

•Simple, fast web access



These data-sets go into the creation of high fidelity maps with real-time change data and high density road coverage on a city wide scale.


Data sets can regenerate within minutes from ongoing scanning to reflect navigation-critical events, like construction work and traffic signals, with pinpoint accuracy. UMO’s mobile mapping system therefore provides an accurate reference point that acts as the basis for effective computer vision and safe autonomous driving.


◼︎

Post-processing


Formulate 3D map with GPS, road and other data sets


HD maps are the foundation on which future mobility is built, and a necessary prerequisite for the most critical safety technologies of driverless autonomy. These maps are essential for autonomous vehicles as they provide redundancy and predictive value to their decision making with real time updates and scalable data management.


UMO’s maps are currently production-ready and capable of enabling level 4+ SAE automation, enabling full driverless operating across all driving environments in a given domain.


Collected Data is processed to provide high fidelity 3D maps and automatically translated into the desired map schema. Processed data is delivered platform agnostically avoiding proprietary data format, integration, latency and lock-in complications.



Data type 1. Point-cloud data


 

HD Map Key AV Functions:


Localization – Where am I?

HD maps allow autonomous vehicles to locate themselves precisely in challenging operating conditions in order to safely navigate the built environment.


Perception – What am I seeing?

HD maps verify real-time sensor readings against an objective measure to confirm navigation options.


Planning – Where to next?

HD maps help autonomous vehicles better understand the built environment enabling them to make better informed and better predicted navigation decisions.

Data type 2. Vector data


 

 

◼︎

Result



 

 

*Visualized example of UMO’s HD maps

◼︎

Planning



Define target area


An area for geofencing is selected and surveyed to determine optimum functionality at deployment sites.


Potential sites are evaluated based on long term strategic goals with a focus on partnerships and scalable deployments.

 

UMO engages in a comprehensive and collaborative planning process and involves actively partnering with private and public sector bodies to guarantee deliverables and successful deployments.


UMO aims to facilitate interactions between, creators, corporations and institutions to align interests multilaterally to the benefit of all stakeholders.

 

 

 

 

◼︎

Acquiring GCP


Generate GCP (Ground Control Points) in geo-fenced area and perform precise surveying


Ground Control Points (GCPs) are defined as points on the surface of the earth of known location used to geo-reference imagery.


Surveying Pyeonghwa-ro in Jeju-Do, South Korea


 

 

Surveying Pyeonghwa-ro in Jeju-Do, South Korea


 

 

Surveying Pyeonghwa-ro in Jeju-Do, South Korea


 

 

GCP’s are the anchor points in which a geo-fenced area is precisely defined and referenced


◼︎

Scanning



Operate mapping vehicles with mobile mapping system installed


UMO utilizes fleet based LiDAR sensors to collect a broad variety of data on the built environment, including:

•Roads: Lane position and regulation, Road boundaries, stop lines, centerlines etc.

•Infrastructure: median strips, tunnels, bridges, underpasses, traffic signs, traffic lights etc.


These sensors operate continuously to provide precise measurements foundational for effective HD maps.

UMO’s mobile scanning process simultaneously collects 3D spatial and locational navigation critical data:

•Centimeter accurate 3D models

•Panoramic street imagery

The following raw data is formulated into highly versatile base maps, that when coupled with UMO’s change-management solution, allows for a wide range of future mobility applications within geofenced areas

 

 

 

◼︎

Scanning



Collect raw data of roads with mobile mapping system


UMO’s fleet based scanning process contextualizes collected road data for the purposes of autonomous localization, perception and planning via:


•Real-time GPS tracking

•Vehicle safety analytics

•Trip video archive

•Simple, fast web access

These data-sets go into the creation of high fidelity maps with real-time change data and high density road coverage on a city wide scale.

Data sets can regenerate within minutes from ongoing scanning to reflect navigation-critical events, like construction work and traffic signals, with pinpoint accuracy. UMO’s mobile mapping system therefore provides an accurate reference point that acts as the basis for effective computer vision and safe autonomous driving.

 

 

 

 

 

 

◼︎

Post-processing


Formulate 3D map with GPS, road and other data sets


HD maps are the foundation on which future mobility is built, and a necessary prerequisite for the most critical safety technologies of driverless autonomy. These maps are essential for autonomous vehicles as they provide redundancy and predictive value to their decision making with real time updates and scalable data management.


UMO’s maps are currently production-ready and capable of enabling level 4+ SAE automation, enabling full driverless operating across all driving environments in a given domain.


Collected Data is processed to provide high fidelity 3D maps and automatically translated into the desired map schema. Processed data is delivered platform agnostically avoiding proprietary data format, integration, latency and lock-in complications.

Data type 1. Point-cloud data


 

Data type 2. Vector data


 

 

HD Map Key AV Functions:


Localization – Where am I?

HD maps allow autonomous vehicles to locate themselves precisely in challenging operating conditions in order to safely navigate the built environment.


Perception – What am I seeing?

HD maps verify real-time sensor readings against an objective measure to confirm navigation options.


Planning – Where to next?

HD maps help autonomous vehicles better understand the built environment enabling them to make better informed and better predicted navigation decisions.

◼︎

Result

 

 

*Visualized example of UMO’s HD maps