Crafting a New Classification Function for Passive Safety

Vehicle Occupant Sensing

Crafting New Capability To Classify

for Passive Safety

Introduction: Classification Function – General Discussion

Continuing exploration of the Occupant Sensing System, let’s delve into the most complex function: “Classification”. Various Governing Passive Safety Features would utilize the comprehension of the size related to the present and seat occupant to apply to their respective functions.

For example, the Notification Features would manifest adults and children, the Restraint Features would advise the airbag deployments, and finally the Collision related Features would request Crash Management System to make an accurate decision while deploying the airbag properly. In addition, there are other possible applications in the comfort and health related functions / features.

Hence, properly classified present Occupants would benefit in many ways from Passive safety Features, while utilizing the vehicle across driving scenarios.

The Significance of the Classification Signal Outputs for the applications

Fundamentally, after some amount of Alive Object’s attributes (data) is collected and processed, there is an opportunity to make a qualitative assessment of the present being and what this data is good for.

In addition, within the realm of Systematic development, it is a proper moment to ask the following question: Why are we collecting this data? What is this data going to be used for? The Systems Engineering Method is applied in Top-Down development of the requirements, meaning the specific Governing Feature needs to be driving the specific requirements to employ the Signal Output for their respective applications.

Therefore, every “interested” requesting Feature shall define their functional requirements and ask for specific output from the Occupant Sensing System. In case of the “classification” function this becomes quite different for each feature.   

Following with an example for the simple presence, we already addressed basic classification in previous article. Consequently, the simple classification is based on the Minimum Criteria Set, and is not enough to discern whether the occupant is an “adult” or a “child”. Therefore, to classify as such the requirement sets need to be more comprehensive.

Continuing with examples, to address the seat comfort functions, the requirement data sets needs to collect information related to the specific body parts contact with the specific seat areas, like seats, arms and head rest.

Finally, and the most complex, in order to properly deploy the airbag, the Crash Management System would like to understand the size, position of the body and even the position of the limbs with relationship to the Airbag.

 

Classification function - Definition

Let’s elucidate the concept of “Classification.” In the fusion of two converging notions—the diverse requirements stemming from various applications and the escalating complexity of data sets—it becomes imperative to establish a sequential development from simplicity to complexity.

Thus, the “Classification” function entails evaluating the collected data and assessing the processing algorithm’s capability to align the data set with the specified requirements for the applications of Governing Features.

As the most intricate among Occupant Sensing functions, Classification follows the initial data acquisition from computations like Occupant Presence Detection, Occupant Location, and Seat Belt Monitoring. It emerges as a follow up of them all and a derivative of the foundational “basic classification” established during the initial presence determination phase.

Approach to the “Classification” Algorithm - Deeper dive into the Verification Cycles

Following to our discourse on Seat Belt Monitoring algorithm development, the journey toward refining the Classification algorithm persists until all prerequisites are satisfied. Consequently, optimal data acquisition occurs once the occupant is securely seated, allowing for assessments in a static condition. In addition, as data accrues over time, the computational algorithm facilitates comparisons against defined requirement sets.

Measurements of the occupant’s dimensions, seat belt routing, and initial seat condition contribute to this evaluation. Furthermore, even when occupants are not properly positioned, certain assessments remain feasible based on established requirements. Moreover, it’s crucial to underscore the sequential logic underpinning data accumulation, analysis of data maturity, and incorporation of new attributes into existing conditions. This sequential approach fosters heightened reliability in Signal Output readiness and facilitation upon request.

Moreover, in cases of power loss or system reinitiation, the computational unit retains the “last known” status, ensuring continuity of computation from that point onward, leveraging the sequential verification algorithm.

Notably, this discussion exclusively addresses “Static” Usecases, where occupants are assumed to remain static within the seat, distinct from “dynamic” conditions where body movement results from unexpected vehicle accelerations or stops—an aspect slated for separate consideration.

Review of the Usecases: Possible Conditions

Additionally, let’s revisit a fundamental use case: the moment when a person approaches and enters the vehicle. Generally, the data acquisition process initiated, capturing the occupant’s attributes right from the start. Sequentially, it’s essential to gather the “points of interest” within the “Field of View” and analyze the available information. Moreover, depending on the technology deployed, preliminary determinations may be possible at an early stage.

Consequently, in cases, where the vehicle is equipped with various sensors, it is important to have all of the “Classification” requirements (Data Sets based on the attributes) available for processing as early as possible for the Vehicle Architecture applications and onboard ECU processing. So, the actual static Human Classification can be determined properly and available for the execution, when necessary.

Therefore, this ensures accurate static Human Classification, primed for execution as needed when the driver assumes their seat and fastens their seatbelt. Concurrently, if a second occupant is present in the second row, their classification should also be completed by this time. Moreover, the verification algorithm assesses door statuses alongside the occupants’ seated, belted, and classified statuses before communicating with the onboard processor.

Additionally, the seat comfort aspect can be concurrently addressed, validating the specific relationships between seat cushions and the occupants’ bodies.

Conclusion: Classification Function – the highest level of Occupant Sensing algorithm, computation and System development

In conclusion, crafting new capabilities for occupant classification marks a pivotal leap forward in vehicle systems engineering. By embracing sensor fusion techniques, machine learning algorithms, and leveraging diverse sensor types, vehicle manufacturers can propel occupant safety to unprecedented levels.

Therefore, enhanced classification capabilities empower vehicles to safeguard occupants across various driving scenarios, thereby fostering safer roads and elevating overall safety standards. In essence, the culmination of the functional verification algorithm lies in achieving comprehensive Classification of the live Occupant, complemented by their properly belted seated position.

Consequently, the system’s ability to furnish timely information to the onboard ECU encompassing all Occupant Sensing functions—including Presence Detection, Location, Classification, and additional metrics like Headcount and Belted status—represents the ultimate objective of technology development.

Furthermore, such sophistication signals the readiness of OEMs and supply chains to deliver vastly improved Passive Safety features for passenger vehicles, including both conventional and autonomous models.

Notably, considerations pertaining to non-living objects and stratification of live occupants’ classification, alongside “dynamic” classification for collision-related scenarios, are addressed separately for further elucidation.

References:

About George D. Allen Consulting:

George D. Allen Consulting is a pioneering force in driving engineering excellence and innovation within the automotive industry. Led by George D. Allen, a seasoned engineering specialist with an illustrious background in occupant safety and systems development, the company is committed to revolutionizing engineering practices for businesses on the cusp of automotive technology. With a proven track record, tailored solutions, and an unwavering commitment to staying ahead of industry trends, George D. Allen Consulting partners with organizations to create a safer, smarter, and more innovative future. For more information, visit www.GeorgeDAllen.com.

Contact:
Website: www.GeorgeDAllen.com
Email: inquiry@GeorgeDAllen.com
Phone: 248-509-4188

Unlock your engineering potential today. Connect with us for a consultation.

If this topic aligns with challenges in your current program, reach out to discuss how we can help structure or validate your system for measurable outcomes.
Contact Us

Leave a Reply

Your email address will not be published. Required fields are marked *.

*
*
You may use these <abbr title="HyperText Markup Language">HTML</abbr> tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Skip to content