Crafting a New Dynamic Classification for Passive Safety #2
Crafting a New Dynamic Classification Function
Introduction - Dynamic Classification Function – Discussion Continued
Continuing the discussion on the complexity of the function: “Dynamic Classification” with the straightforward frontal collision Usecase. Further addressing the development of the requirements necessary for the Data Acquisition and Processing for the Signal Output.
Fundamentally, the developed prerequisites are covered by the Occupant Sensing functions related to the “Static” Usecases. This encompasses all of the occupant’s behavior prior to the final seating and belting, and assumes the static body position with respect to the seat.
Consequently, in the event of the unexpected vehicle actions, such as sudden stop, deceleration and or an actual collision, the body becomes “Dynamic” and moves, including parts of the body, such as head and limbs. Basically, the advantage of having technology capable of the “Dynamic Classification” is an early signal output to the Airbag deployment Management function, which will help to manage the timely deployment of custom (smart) airbag for a specific body size.
Sequentially, this article delves into the “Dynamic” Frontal Collision Usecase development, looking into the definition of the dynamic classifier criteria (requirements set).
General Discussion – Sudden Change in “Vehicle in Motion” Condition
Firstly, let’s discuss the very general topic of the “Vehicle in Motion” vs. the sudden change. Consequently, the Vehicle would transition from the normal condition to the abnormal. Therefore, the driven vehicle will have to suddenly decelerate for whatever reason related to the external phenomena. Moreover, this can happen from aspects related to the driver losing attention to the instantaneous changes to the road and traffic conditions. Then, the possibilities are endless.
Sequentially, what constant at the moment (from Occupant and Vehicle Safety standpoint) is the vehicle size (external and internal) and occupant general characteristics, i.e. attributes. Furthermore, this means that occupant’s body potential movement is limited to the Interior boundary (condition) and Seat Belt, if used. In addition, there is also a possibility of the going straight through the windshield. Hence, this specific Usecase can be modeled by the applicable tools, as only specific geometry, collision conditions and body sizes – will generate such result.
Consequently, the Vehicle abnormal conditions maybe classified as the following:
- Firstly, false scare – the event when reasonably on time the vehicle became stable and in control (in motion), resumed the normal driving; conditions in which the potential dynamics of the occupant’s body motion is miniscule
- Secondly, Actual scare – the event when certain road or traffic conditions change to create uncontrolled vehicle motion, however, with help of certain vehicle functions the vehicle resumed the normal driving – conditions in which the potential dynamics of the occupant’s body motion warrants preparation of the Airbag deployment, and after the assessment deemed unnecessary
- Finally, actual collision – the event in which the vehicle ended up in an actual collision, with occupant’s body dangerous displacement, and would require the Airbag deployment
General Discussion – Occupant’s Body Condition
Generalizing the Occupant’s behavior, let’s present two major body conditions, which will require assessment, and further actions:
- Occupant is properly seated and properly belted. In this condition, all of the “Static” algorithm is possible with a complete assessment of the body size and potential motion during the abnormal Vehicle event. This is an ideal situation for the Occupant Safety. Please, wear your seat belt
- Occupant is seated and not belted. In this situation, dependent on the body position and time allowed for the computation, the body size can be approximated. Also, with the developed algorithm the maximum potential travel / shift of the body can also be assessed / predicted. (This is projection of the future development, although possible and feasible.)
Consequently, these are aspects of the Occupant’s behavior to account for in case of the Vehicle abnormal Usecase.
Discussion: the “Dynamic Classification” Governing Algorithm
Looking at the Vehicle level function (Collision Occupant Protection) to assist with Occupant Safety during the actual collision, the following must be considered for the development of the governing Algorithm:
- The vehicle behavior types – normal and abnormal (discussed above)
- The Occupant behaviors – allowing maximum safety and computations related to the “Static” Usecases, as well as conditions in which the seat belt is not used. Consequently, this would require an additional predictive computation
- The total (top-down) sensor logic, accounting for all of the incoming information (if equipped) from the external sensing functionality and necessary internal sensing capabilities
Generally, the development of such technology would require the Vehicle level Feature mapping of all of the participating interfaces and logic for the proper integration.
Dynamic Classification function – Definition – Deep Dive Clarification
Basically, for the Usecases in which the occupant is belted, the possibility of the body parts displacement is less and can be modeled based on the known vehicle and test data, some particular collision instances can be always added to the appropriate simulation tools and models. The outcome of this study would create comprehension related to the Usecases – quantifiable potential displacement.
Consequently, the more complex development is necessary for the unbelted occupant’s body displacement, in order to determine the danger level and specific travel during the predicted collision events.
Furthermore, let’s define this hypothesis – What is the “Dynamic Classification” concept?
Defining, Dynamic Classification (quantified) is the capability of the Occupant Sensing System to measure (in timely manner) the amount of Occupant’s body potential displacement from the initial position (within the Interior) caused by the change of the Vehicle behavior, before crossing the already defined threshold, to inform the Airbag decision making ECU, in order to manage the appropriate airbag deployment.
Rephrasing / clarifying: Whether it is a head motion, or a hand flying – the quantifiable part of the body measured within defined spacial and timing brackets, then communicated to the decision-making unit – what constitutes Dynamic Classification function (algorithm)
Defining the “Dynamic Classifier” for the Frontal Collision Usecase
Generally, it is worth mentioning that not too many collisions happen instantaneously. In plurality of the scenarios both vehicles are in the proximity of various (potentially available) sensor technologies. Consequently, this means that the initiation of the “collision protocol” can come way before the actual event, thus allowing ample time to verify the initial position of the body, belted status, potential opportunities for body travel, other factors.
Furthermore, this time allowance is a lot longer than the period from the initiation of the body jerk to the body end of travel in the collision. Hence, this allows for the necessary logic for computations and verification, providing the signal output before the body crossing of the predetermined threshold. The distance between the imaginary border to the A-Side of the Instrument Panel needs to be determined to allow for the airbag deployment time and space.
Finally, defining the “Dynamic Classifier” as the set of conditions accounted by the governing logic and communicated to the decision-making unit, based on the computed volume of the body element or elements deemed to be sufficiently large to cause the occupant’s body harm.
Moreover, this is derivation made from the initial “Static” body position computation and the predicted trace of the potentially moving body parts, collected between the initial body position and certain predetermined plain or length of motion curve. Additionally, all data acquisition is based on the originally established Occupant’s attributes – for the Dynamic Classification – the algorithm account for both: known preliminary computed data set and collected dynamic information during the vehicle event.
Conclusion: Dynamic Classification Function Development
In conclusion, the following are relevant notes related to the development of complex systems for the Occupant Protection purposes, in this case Dynamic Classification Function:
- Firstly, the Usecase assumed is the Frontal Collision impact – the simplest approximation for the purpose of predicting the Occupant’s body motion
- Sequentially, all other Usecases would have to have a similar approach accounting for the direction of the impact and other possibilities related to the body motion
- Then, the autonomous vehicles will have to have the same approach for the same Occupant Protection purposes, as there is no driver present in the vehicle
- In addition, there is a need to develop a set of the simulation tools for the development / verification and even validation of various elements of this technology. Moreover, Virtual Reality and AI tools can be extremely helpful in the necessary phases of this development
- Notably, both occupants’ conditions (belted and not belted) need to be considered for the algorithm development
- Moreover, the specific requirements for the Dynamic Classifier (s) will depend on the scope of (Vehicle Collision related) Usecases and potential opportunities for body (parts) travel within the specific interior
- Finally, the development of the governing algorithm would mean the integration of various Vehicle Systems dedicated to their specific functions and contributing to overall expected level of Occupant and Vehicle Safety
References:
- System Prerequisites https://georgedallen.com/usecases-development-of-the-prerequisites-new-databases/
- Governing Features https://georgedallen.com/development-of-the-prerequisites-new-passive-safety-features/
- Occupant Presence Detection https://georgedallen.com/develop-presence-detection-new-occupant-sensing-tech/
- Occupant Location https://georgedallen.com/develop-occupant-location-new-sensing-tech/
- Basic Classification https://georgedallen.com/develop-the-core-of-occupant-classification-new-sensing-tech/
- Seat Belt Monitoring Function development Part1 https://georgedallen.com/crafting-new-seat-belt-monitoring-tech-for-passive-safety/
- Seat Belt Monitoring Function development Part2 https://georgedallen.com/crafting-new-seat-belt-monitoring-tech-for-passive-safety-2/
- Classification Stratification – https://georgedallen.com/crafting-classification-stratification-new-passive-safety-tech/
- Dynamic Classification – https://georgedallen.com/crafting-a-new-dynamic-classification-for-passive-safety-2/
- Classification of Non Living Objects – https://georgedallen.com/develop-the-core-of-object-classification-new-sensing-tech/
- Seat Belt Monitoring Development https://georgedallen.com/advanced-seat-belt-monitoring-technology-in-occupant-safety/
- NCAP Standard https://www.euroncap.com/en/car-safety/the-ratings-explained/safety-assist/occupant-status-monitoring/
- Seat Belt Wikipedia: https://en.wikipedia.org/wiki/Seat_belt
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.
