The spread of autonomous driving requires resolving the lack of knowledge and resources

As autonomous driving becomes more widespread, the demand for LiDAR sensors and the software required for operation is also increasing. However, autonomous driving technologies are diverse, and many companies lack the knowledge and resources to develop LiDAR sensors and the associated software such as "Perception" and "SLAM."

In addition, it seems that the characteristics of LiDAR sensors vary by manufacturer, and the costs involved in implementation also differ. Even when comparing sensor specifications, there are many items to compare, making it very difficult for a single company to select the appropriate product for its purpose and application.

Furthermore, it seems that the officially announced specifications of LiDAR sensors may differ from the specifications obtained by actual measurements on site. Because the prerequisites for the on-site usage environment, climate, temperature, etc. are different, the values written on the spec sheet do not necessarily match the actual measured values.

For these reasons, LiDAR sensors must be purchased at one's own expense and their performance evaluated and verified, which would be extremely costly and time consuming. Even if one wants to introduce autonomous driving, there is a lack of knowledge and resources, and if there is no ample budget to allocate to autonomous driving, it is difficult to even evaluate and verify LiDAR sensors. The current situation is that verifying the performance of sensors at one's own expense is difficult in terms of both budget and time.

So this time, we interviewed Macnica, a company headquartered in Kohoku-ku, Yokohama, which provides support for autonomous driving. We visited the garage, which is the base of technology development for the company's Smart City & Mobility Division, and spoke with the person in charge about their revolutionary service, "Sensor Consulting," which helps solve the issues surrounding autonomous driving. We also had a demonstration of Sensor Consulting at the garage.

What is Sensor Consulting, which supports product selection and implementation?

Sensor Consulting is a service that provides consistent support from the selection to the introduction of sensors related to autonomous driving. As mentioned above, there are many types of LiDAR sensors, Perception sensors, SLAM sensors, etc. required for autonomous driving and ITS, making it difficult to select.

A representative said, "Sensor Consulting examines the customer's application and usage environment, selects sensors and software that fit the conditions from a huge number of options, and supports them through evaluation and implementation." Sensor Consulting not only selects sensors, but also evaluates and verifies the sensors at the actual installation site, providing support right up to implementation.

Macnica The person in charge explains that there is a good reason why the company is able to consistently support the implementation of sensor technology used in autonomous driving and other products.

"the Company have a wealth of knowledge and resources. Since 2017, we have been one of the first to focus on autonomous driving technology, and have been discovering, investigating, and studying various technologies emerging around the world. We have knowledge and experience of all autonomous autonomous driving technologies, and have a range of appropriate products (hardware and software). We have a system in place that allows us to support companies with issues in product selection, evaluation, verification, and system construction."

Macnica is not only able to help companies reduce their time to market (TTM) and alleviate resource shortages, but it also has a sourcing function that uses its advanced information gathering capabilities to select and negotiate with target companies, development support with experienced engineers, and a service delivery structure that cannot be achieved by its competitors, which has enabled it to provide a unique sensor consulting service.

Sensor Consulting Flow

Sensor consulting will mainly be carried out based on the following seven items. However, as it may vary on a case-by-case basis depending on the application and purpose, please use the items as a reference only.

1. First hearing

We ask basic requests, such as what the sensor will be used for, the environment in which it will be used, and its scale, and then select several products, including sensors, that seem to fit the purpose.

2. Requirement definition

We then determine detailed requirements, such as product specifications, to see whether the sensor selected in the initial interview is suitable for the usage environment.

3. Sensor selection based on desk-based analysis

Narrow down the most suitable sensors based on product specifications.

4. On-site evaluation of selected sensors and evaluation plan

The selected sensors will be evaluated on-site to see if they can be used properly in the actual environment, and an evaluation plan will also be created.

5. Conduct on-site evaluation of selected sensors (on-site inspection, rehearsal)

On-site inspections and rehearsals are carried out to assess whether the sensor can be operated properly. Careful on-site assessments are essential because the sensor may not perform as specified depending on the environment.

6. Preparation of evaluation report

An evaluation report is prepared based on the on-site evaluation. Depending on the situation, an on-site evaluation may be conducted again.

7. Reporting session - sensor delivery

Based on the evaluation report, we hold a debriefing session with the relevant parties, and if it is deemed to be properly operated, we deliver the sensor and associated products.

Examples of sensor consulting examples and data obtained during comparison

Here we will introduce examples of sensor consulting that Macnica has actually carried out. However, the numerical values have been changed when publishing these examples. Please use them as a guide to what kind of data can be obtained and how it was carried out.

[Example of evaluation report]

Data provided by Macnica

Data provided by Macnica

As an example, a comparative study was conducted to see what the differences are between LiDAR sensors and stereo cameras. In addition, a LiDAR sensor and a stereo camera were installed on a high-speed vehicle traveling at 60km / h to see how much difference there is in the data that can be acquired. Furthermore, they are also conducting studies to see how much the data that can be acquired is affected by bad weather (special environments) such as rain, fog, and dust, and to verify the accuracy of LiDAR SLAM on large vehicles with a total length of over 20m. By conducting studies and evaluations to introduce sensors that are suitable for the situation and purpose, they are able to propose appropriate products that meet customer needs.

[Example of SLAM verification (accuracy verification)]

Photo: Courtesy of Macnica

[Example of LiDAR SLAM verification in a wide environment]

Photo: Courtesy of Macnica

LiDAR is expanding its use beyond automobiles

The performance of LiDAR sensors has improved as the development of autonomous driving technology has accelerated. At the same time, costs have fallen, and LiDAR sensors are increasingly being used in applications other than automobiles.
By using LiDAR sensors instead of cameras, it becomes possible to grasp the positional relationship with the target object in detail without being affected by light such as surrounding lights. In addition, since the face is not captured in the image, it has the advantage of being able to respect privacy, and the range of applications is expanding.
There is also an accelerating trend to apply LiDAR sensors in the security field, such as by using them as a replacement for surveillance cameras, or to analyze the flow of people in public places and traffic volumes on roads.

Experience sensor consulting for free

Macnica offers a "free sensor consulting trial service" that allows customers to experience sensor consulting free of charge according to their needs. This service starts with an initial consultation, and then Macnica staff visits the site with sensors, allowing customers to see for themselves what kind of data can be obtained in their environment.

We had a demonstration of an on-site evaluation using a LiDAR sensor, demonstrating how accurate information can be obtained both day and night.

The person in charge said, "If you are having trouble selecting products such as LiDAR sensors, associated software, and peripheral devices, please feel free to contact Macnica with our extensive knowledge." Macnica has a good reputation for product selection and has a long track record of implementation. This is a service that companies considering introducing autonomous driving should take advantage of.

Text and photos: Yamato Muroi (freelance writer)