Solution > MaaS-Korea Corp



Road Classification

We examine our community roads and classify them according to their autonomous driving safety to learn where to first deploy autonomous vehicles.

• We Investigate and evaluate various risks related to the operation of autonomous vehicles in our local transportation infrastructure and assign autonomous vehicle safety grades to all roads in our community in tables and maps.

• We evaluate the road infrastructure factors in our region through literature and field surveys. We collect data on autonomous driving environment of our region repeatedly by utilizing our own vehicle equipped with 64ch LiDAR and IMU and our own Drone equipped with survey sensors. We process the collected data for our artificial neural network to give the autonomous vehicle safety rating for each road segment in the region.

• We help to decide which services to start with, with which vehicles, in which regions, and create a specific introduction roadmap.

CAV Impact Analysis

We present appropriate business models for autonomous driving service in our region and analyze the effects of introducing autonomous driving services with traffic simulation models.

• We design autonomous driving service suitable for our region and present business model. The economic and non-economic effects expected from the introduction of autonomous driving services are presented independently and objectively from the autonomous vehicle manufacturers.

• We design services and derive business models with established marketing research and service design methodologies to reflect the unique needs of our consumers in our region.

• We create traffic simulation models for our region to estimate the effect of the introduction of autonomous driving service on our local transportation quantitatively. This will help increase the residents' acceptance.

Master Planning

We provide a master plan for introducing self-driving services tailored to our region. We present practicable self-driving vehicles, service areas, service contents and effects, and mid- to long-term roadmaps.

• We present a blueprint for autonomous vehicles introduction to our region and longer-term autonomous driving services expansion plan.

• We comprehensively analyze the safety level of autonomous vehicles and the effect of introducing autonomous driving services on our local roads. Based on the results, we compare and analyze the autonomous vehicles that can be introduced in our region. We design autonomous vehicle service routes and provide future expansion strategies.

• We help make reasonable investment decisions for local governments, organizations and companies to introduce autonomous driving services.

Accident Prone Spots

We utilize artificial intelligence to identify spots where ADS are more prone to safety accidents.

• Based on our data analysis we suggest ways to improve safety by finding the spots of high accident probability related to autonomous driving.

• We analyze the autonomous vehicle specifications, sensors, sensor fusion and safety-related software algorithms. In addition, we conduct a detailed survey on the operating section of autonomous vehicles using equipment such as our survey vehicles and drones for the driving environment. We identify autonomous accident-prone spots through repeated analysis of various ODD factors. The results of the analysis are used to suggest ways to improve accident prone spots through collaboration with local governments and infrastructure design engineering companies.

• Autonomous driving routes can be improved to prevent accidents and to improve the acceptability of residents.

In situ ADS Test

We test and validate the safety performance of the autonomous vehicles to be introduced at the identified accident-prone spots in accordance with international standards.

• Autonomous vehicles equipped with AI-based ADS cannot be tested completely for their safety with traditional test methods consisting of simulation, closed track test and open road test. We test the autonomous vehicles in accordance with international standards at the locations identified as our accident-prone spots in our region and verify the safety.

• To ensure reproducibility, reliability, and practicality of tests, we design and execute in accordance with international standards. Public or closed tests and verification by an external accreditation organization ensue.

• Community acceptability can be enhanced by providing concrete evidence to our stakeholders. In addition, we can improve the performance or specification of autonomous vehicles that are found to be less-than-perfect through our testing and verification.

ODD Expansion

We make it possible to expand the time and area for autonomous vehicles to drive safely, increasing the affordability and acceptability of autonomous driving services.

• ODD is the set of driving conditions under which an autonomous vehicle can operate safely. In order to increase the daily operating hour of autonomous vehicles and increase the service area, the ODD must be expanded. We analyze the sensors, communication devices, perception algorithms, operation devices, controls, and HMI of the self-driving vehicles to identify the types of failures associated with safety and service levels through FMEA methodology and optimize the ODD in our local climate and with the transport infrastructure. We conduct a variety of ODD scale-out tests at our local sites to attain the best ODD.

• With optimized ODD, daily operating hours and service areas can be expanded. Autonomous driving services can be more viable economically.

High Currency LDM

We detect changes in safety related roadside assets and updates LDM by reflecting the changes up to minutes.

• Concept: Local Dynamic Map is a collection of location-based data. It is a synthesized information on all objects, including road shapes, road signs, vehicles, and pedestrians, presented onto linked layers. HCLDM is a dynamic map with high currency of which information in the upper LDM layer is updated within minutes.

• We classify and register all roadside assets that affect autonomous driving safety in our area and update them in HCLDM by detecting changes. HCLDM is an essential input to the real-time autonomous driving safety assessment algorithm. For significant risks of detected changes shall be reflected in the dynamic ODD.

•Through this, it is possible to prevent autonomous vehicle accidents caused by sudden changes in roadside assets.

Real-time ODD Feed

We assist safer autonomous driving by providing the optimum ODDs that reflect real-time information such as weather, traffic control zones, construction area, and disaster alerts to the ADS without delay.

• ODD are defined as complex interactions between various ODD factors including vehicles, pedestrians, roadside assets, physical infrastructure, traffic laws, connectivity, environmental conditions, zones, construction areas among other things. Many ODD factors change in real time, and the magnitude and speed of change are sometimes unpredictable. In order to support safe operation of autonomous driving vehicles, ODD must be updated in real time to reflect dynamic ODD factors and real-time regional information such as disaster alerts, geofencing set by local governments, demonstrations and other relevant events and then be provided to ADS.

• We provide real-time ODD OBU that feed real-time ODD to ADS. The specific feedback structure is customized according to the specifications of sensors, communication devices and perception algorithms of autonomous vehicles. Our ODD OBU notifies ADS when an autonomous vehicle is out of the ODD range to assist in initiating a safety stop or a fail operation to attain minimal risk condition.

• It is possible to reduce the accident rate that can occur by preventing the autonomous vehicles from continuing to drive in the sudden harsh weather conditions and in other unexpected situations.