In 1969, Frank E. Bird conducted a study for the Insurance Company of North America, covering 1,753,498 accidents reported by 297 companies in 21 different industrial groups, representing 1,750,000 people and 3,000,000 hours worked.

Out of 641 abnormal situations:

• 600 are near misses/incidents;
• 30 are property damage accidents;
• 10 are minor injuries;
• 1 is a serious accident.

See Frank E. Bird Risk Pyramid in image below.

The pyramid representation expresses the fact that the most serious accidents are the rarest and that, conversely, the abnormal but benign situations are the most frequent. This representation is symbolic, as the width of a level is not proportional to the frequency (unlike a diagram), the pyramid merely shows a ranking. However, it also expresses the fact that the difference between a harmless and a serious situation is often only due to luck: if an accident is narrowly avoided, it has no consequences, but the difference between harmful or even serious consequences is very tiny. The ratio of one level to the other of the pyramid gives an estimate of this luck factor, while the size of the base of the pyramid gives an estimate of the determination factor.

During its road safety research work (12 funded collaborative scientific programs), NEXYAD applied this study to road risk. Indeed, road accidents are ultimately rare in relation to the overall mileage and very difficult to analyze. On the other hand, experts and the road police analyze the footage from surveillance cameras installed here and there on the road infrastructure, which report abnormal or critical situations that are – much more numerous – but do not lead to accidents. These near-accidents are almost always the result of a lack of anticipation and prudence on the part of one or even several drivers in relation to a particular context. Being without consequences, they do not appear anywhere in public body and insurance statistics. More seriously, in the absence of consequences, drivers at fault may not be aware of the risks they have taken.

In this context, NEXYAD has developed SafetyNex a metric of driving prudence that detects both accident conditions and the usual lack of prudence where we are lucky… but luck can turn, and the more reckless we drive, the more we statistically increase the bad luck of having an accident.

SafetyNex is an AI based on the knowledge of road safety experts to measure prudent driving and detect risky behaviors. This tool is available for fleets, insurers and driving schools in a BYOD version, and for automated and autonomous driving of production vehicles.
For the BYOD version, the objective is to monitor the driver directly in terms of prudence, rather than calculating indirect estimators such as harsh braking, for example, which are little or not correlated in practice with accidents. We are applying this with the Indian company Motiv.AI, which has developed a smartphone app based on our technology and is deploying it in India for fleets.
For production vehicles, it is first a question of measuring their performance in terms of caution in autonomous or automated mode, and then simplifying the driving models (drastically reducing the number of parameters) because it turns out that our caution functions describe the problem of automatic driving according to a formalism of independent variables. This results in a reduction of at least a 100 factor in development and testing times. We have signed a partnership agreement with the company STELLANTIS on this subject.

See https://nexyad.net/Automotive-Transportation/intelligent-mobility/

#ADAS #AutonomousDriving #HAV #AutomatedDriving #RoadSafety #AI #ArtificialIntelligence #RiskTriangle #FrankBird #SafetyNex #Nexyad #SmartphoneApp #Motiv.AI #India #Stellantis

St Germain en Laye, September 18th 2024.

Interesting Podcast proposed by PennDOT and The Spark:

« Over the summer PennDOT opened to the public a comment period on the Highly Automated Vehicles (HAVs) guidelines. The HAV advisory Committee was created in 2018 to advise and consult the Secretary of Transportation on each aspect of highly automated vehicles.

Derrick Herrmann, Chief of Transformational Technology at PennDOT, and Matthew Johnson-Roberson, Director of the Robotics Institute at Carnegie Mellon University joined The Spark, to share more about research about the HAV and driverless vehicles »

#TheSpark #PennDOT #CarnegieMellon #DerrickHerrmann #MatthewJohnsonRoberson #Driverless #HAV #ADAS #AutomatedDriving #Nexyad

St Germain en Laye, september 16th 2024.

SafetyNex by Nexyad is an ADAS tool that measures prudent driving on a scale of 0 to 100 (100 = maximum prudence and 0 = maximum risk).

This is compliance with the road traffic code and the practices of experienced drivers. Trivial examples: we do not brush against pedestrians walking on the side of the road, we slow down before an intersection, we do not run stop signs, we adapt our inter-distance with vehicle ahead, etc. This compliance with the rules is reflected in the anticipation of complex situations and a speed adapted to each context. Since zero risk (or lack of prudence) does not exist, it is up to the driver or the driving robot to minimize their own risks as much as possible and to anticipate possible lack of prudence by other users.

For automated/robotic driving systems, such as predictive ACC, and L3, L4, SafetyNex is set to a given prudence threshold based on the information available at any time. If the vehicle is omniscient because it is equipped with many perception organs and the road infrastructure also sends it information that is undetectable by itself, then with full knowledge of the facts the prudence threshold can be lowered. On the other hand, if the information on the road context is incomplete (for example, perception does not cover all angles around the vehicle) then the prudence threshold must be raised so that the speed is reduced a priori in situations of contextual uncertainty.

Let us recall the definition of prudence: taking into account in one’s behavior that one cannot predict or know everything, and that one cannot predict all the consequences of one’s actions.

Concrete example:

SafetyNex is an AI, based on human knowledge, resulting from 15 years of R&D and 12 collaborative scientific programs with road safety experts, road police, professional drivers and car insurers.

See NEXYAD page: https://nexyad.net/Automotive-Transportation/prudence-based-automated-driving/

#Nexyad #SafetyNex #ADAS #AutomatedDriving #AutonomousDriving #RoadSafety #AI #ArtificialIntelligence #PredictiveACC #Level3 #Level4 #DrivingPrudenceMeasurement

St Germain en Laye, September 13th 2024.

It is striking that professional fleets and car insurers rely on indirect indicators of harsh driving to monitor their drivers and assess their driving. These indicators are poorly correlated with accidents and near-misses.

Why talk about near misses? Because they are much more numerous than accidents (easier to analyze) and meet all the conditions of an accident – except one. For example, if one accelerates at the intersection of a priority road, he/she has a dangerous driving, but accident happens only if a vehicle crosses at the same time. If there is no vehicle, driving remains dangerous but you are « lucky » and the accident does not happen.

It is to take this phenomenon into account that we have developed at NEXYAD a metric of prudence that detects any reckless behavior whether there is an accident or not. The explicit calculation of driving prudence simplifies interpretations, is useful for training drivers and for monitoring their progress.

The first person to whom one sends his lack of caution is the driver himself by an alert triggered a few seconds before if and only if his behavior is reckless. This is the direct and instant prophylactic effect. The impact on accident rates reduction is 20% to 30% (reduction in the number of accidents in an equi-distributed manner; standard accidents, serious accidents, fatal accidents).

This metric of prudence is available as a BYOD (Bring Your Own Device) as a ready-to-use smartphone app or as an SDK for IT developers.

See the NEXYAD solution: https://nexyad.net/Automotive-Transportation/byod-solutions-for-fleets/

St Germain en Laye, september 10th 2024.

This multi-billion dollar bet – and the colossal investment is not over yet – aims to eliminate the driver from the cab equation while ensuring 100% safety during the ride. The driver’s salary is indeed a significant cost factor: around 150,000 dollars per year per car for 24/7 operation.
Today, the profits are not there yet, but huge US companies like Waymo, GM or Amazon are pushing for it because they believe in it.

« Waymo’s Robot Taxis Are Almost Mainstream. Can They Now Turn a Profit?
Now that its technology is showing it can work on city streets, Alphabet, which owns Waymo, plans to invest billions more.
Waymo is now completing over 100,000 rides each week in San Francisco, Phoenix and Los Angeles — double the number in May. And on a July earnings call, Alphabet executives said the company planned to invest an additional $5 billion in Waymo. »

Read article by Eli Tan for The New York Times: https://www.nytimes.com/2024/09/04/technology/waymo-expansion-alphabet.html

Nexyad offers a hybrid physics-informed AI, on-board and real time, as a safe driving layer for autonomous vehicles, providing a safety reference to industrial players, insurers and public regulatory bodies: Autonomous Cars & Autonomous Trucks driven by Prudence (nexyad.net)

#Nexyad #SafetyNex #ADAS #AutomatedDriving #AutonomousDriving #RoadSafety #AI #ArtificialIntelligence #RobotTaxis

St Germain en Laye, September 5th 2024.

Interesting article, worth to read, by Richard Bishop for Forbes: What’s the Future of Driverless ? Here’s what Experts had to say

The article provides an overview of the 13th Automated Road Transportation Symposium (ARTS24), which was held at the end of July and brought together 600 experts in autonomous mobility.

At present, this mainly concerns the USA, which is the leader in the field of private and public experimentation, and to a lesser extent China. But driverless mobility could follow with few years delay in Europe, and rest of the world.

Richard Bishop overview:
Late this year and early 2025 are shaping up to be a breakthrough period for commercial driverless operations, building on the impressive level of deployment Waymo has already achieved.
While U.S. regulations are unclear, startups have what it takes to reach the ultimate goal of profitability.
Overseas deployments are also expected to make headlines next year.
As this new technological capability matures, a new wave of technological improvements could emerge.

Read Article by Richard Bishop : What’s The Future Of Driverless? Here’s What Experts Had To Say (forbes.com)

Nexyad offers a hybrid physics-informed AI, on-board and real time, as a safe driving layer for autonomous vehicles, providing a safety reference to industrial players, insurers and public regulatory bodies: Autonomous Cars & Autonomous Trucks driven by Prudence (nexyad.net)

#Nexyad #SafetyNex #ADAS #AutomatedDriving #AutonomousDriving #RoadSafety #AI #ArtificialIntelligence #PredictiveACC #Level3 #Level4 #DrivingPrudenceMeasurement

St Germain en Laye, September 4th 2024

Generative AI Tutorial Series by Michigan Institute for Data Science
: Using Generative AI to Assist in Coding in Data Science Research

Generative AI is revolutionizing the research landscape by enabling unprecedented levels of automation and innovation, enabling major breakthroughs in all areas of research. To equip researchers with the skills necessary to transform their research with generative AI, the Michigan Institute for Data Science and the Michigan AI Laboratory collaborated in fall 2023 to offer a series of tutorials consisting of lectures and hands-on demonstrations by U-M experts to train researchers in using generative AI to improve research workflows, coding, creating custom models and more.

See Generative AI Tutorial 1/9:

Nexyad AI Services offer: Artificial Intelligence Services: Solutions to your Practical Problems (nexyad.net)

St Germain en Laye, September 3rd 2024.

A new report from RESEARCH AND MARKETS projects global investment in autonomous driving software to increase by 13.3% per year, from $1.8 billion today to $7 billion in 2035.

« Autonomous Driving Software Market by Level of Autonomy (L2+, L3, L4), Vehicle Type (Passenger Cars, Commercial Vehicles), Propulsion (ICE, Electric), Software Type (Perception & Planning, Chauffeur, Interior Sensing, Monitoring) – Global Forecast to 2035 »

« With the growing advancement and adoption of ADAS features, the Complexity and sugnificance of software in these vehicles is gaining consistent traction. Today, the passenger cars segment has already achieved 45% – 60% penetration in the L2/L2+ segment. As the automotive industry moves towards L3 and above autonomy levels, the cost of software in the vehicle is expected to rise to as high as 45% in the short term and fall consequently by 2035.Futhermore, the proliferation of electric vehicles and shred mobility services is increasing the demand for autonomous driving software. To stimulate demand, the new OEM wants to convert to a software-driven vehicle centralized architecture, develop modular vehicle skateboards, and increase spending on complementary technologies such as AI/Gen AI, intelligent cockpits, and higher levels of automation. »

Report highlights:

– Europe is expected to have significant growth during the forecast period.

Read RESEARCH AND MARKETS report summary: https://www.researchandmarkets.com/reports/5995312/autonomous-driving-software-market-level#tag-pos-12

Read NEXYAD offer for sotfware prudence-based autonomous driving: https://nexyad.net/Automotive-Transportation/prudence-based-automated-driving/

St Germain en Laye, August 29th 2024.

Since the turn of the century, automobile manufacturing has been defined by an assembly of mechanical and electrical functions: an engine, a body, wheels, steering, etc. And different control organs called ECUs (Electronic Control Unit) which manage more and more computer functions: engine maps, braking, windshield wipers, then driving assistance or ADAS and finally infotainment.
Currently, numerous calculators, likely computers, are integrated into vehicles, each responsible for specific functions such as the operation of rearview cameras. Once a function is established and meets the standards of performance and reliability, manufacturers typically refrain from altering it for that particular car model. The process of defining and assembling a model spans approximately 5 to 6 years, during which its functions become fixed and unchangeable.

Tesla revolutionized the automotive industry by creating vehicles based on a general IT framework known as the Software Defined Vehicle (SDV). This concept resembles a computer equipped with seats and wheels. The key distinction lies in Tesla’s ability to enhance its vehicles over time via software updates, in contrast to competitors whose vehicles tend to remain largely unchanged throughout their lifespan. A common scenario involves the necessity of visiting a dealership or service center merely to update the navigation system. Currently, nearly all manufacturers are beginning to investigate this innovative approach, with India appearing to progress at a notably rapid pace.

Today, focus on India which seems to be taking the lead on this issue. Sriram Lakshminarayanan, CTO of Tata Technologies, was interviewed by Techcircle.com. Here some extracts:
“Vehicles are now evolving into high-performance computing platforms, and we’re at the forefront of this transformation. We’re engaged in cutting-edge work, developing multiple proof-of-concepts (POCs) and minimum viable products (MVPs) that explore how this evolution will unfold.”
“Often, IT companies have supported automotive projects in a piecemeal fashion, focusing primarily on areas like mobile app development or cloud hosting. We position ourselves as a full-spectrum vehicle engineering company. We combine our expertise in Full Vehicle Programs with cutting-edge technology, offering a cohesive, end-to-end solution.”

In other words, similar to smartphones, future vehicles will have the capability to support software from various providers, as NEXYAD, to control all functionalities, not limited to infotainment, even post-purchase. The manufacturer will merely need to update the software via over-the-air downloads.

Read complete article here: https://www.techcircle.in/2024/08/28/unlike-it-services-piecemeal-approach-we-focus-on-full-spectrum-vehicle-engineering-tata-technologies-cto

Read NEXYAD offer for SDV, prudence-based autonomous driving: https://nexyad.net/Automotive-Transportation/prudence-based-automated-driving/

Bengaluru, July 7th 2024.

MOTIV.AI India developed the Smartphone App Rudu to alert drivers of professional fleets when lack of prudence exceeds a threshold.

St Germain en Laye, July 3rd 2024.

This is a first result of a tough work since a few weeks, thanks to dSPACE Aurelion Simulation Software and map by HERE Technologies. Nexyad team created a virtual 3D environment from real roads around Paris. Then it is possible to integrate our software product SafetyNex into a virtual vehicle (our democar Dreamotor1) to test use-cases for ADAS and AD systems, in particular here Predictive ACC.

See below 6 differents use-cases around Pedestrian Crossing:

How it works ?

Nexyad works on EVENTS instead of scenarios.
Events are for example: a pedestrian crossing, a stop sign, a curve, a car ahead, a bicycle, etc.
We associate a PRUDENCE FUNCTION to each detected event in parallel. Prudence functions are dynamic as events are.
The combination of detected events make then a “dynamic scenario” that is taken into account FORMALLY by LOGICAL fusion of prudence functions*
Prudence functions are decomposition of driving in independent functions.
By doing so, we do not need an infinite number of  Scenarios to record in a large database. This reduces complexity by a 100 factor for design of Self-Driving tasks. Duration of development, test and validation are reduced with the same rate and computing is much lighter (energy consumption and heat production onboard).

* AI methods used are: Fuzzy Logic & Theory of Possibility

SafetyNex fusions all available informations onboard and computes 20 times per second Prudent Driving Behavior using the Metric of Driving Prudence also developed by Nexyad.

SafetyNex fusions all available inputs information (map, sensors, data streams) to compute Prudent Driving 20 times per second, thanks to the Nexyad Metric of Driving Prudence.

The Metric of Driving Prudence has been developed during more than 15 years thru 12 scientific collaborative research programs with experts of road safety, professional drivers, police of the road and insurers of 19 countries.

Velizy, June 19th 2024.

We had the pleasure to showcase our road safety AI SafetyNex for ZF around Stellantis techno centre in Velizy, with our DREAMOTOR1 demo car.
SafetyNex provides in real time prudent recommended speed, lack of prudence alerts and instant prudent speed for predictive ACC and Autonomous Vehicle.

Thanks to our friends coming from Germany and France.

Paris la Défense, June 4th 2024.

Thanks to all HERE Technologies team for organisation of this yesterday event, which gave opportunity to NEXYAD (We Save Lives) to be up-to-date with automotive trend thru HERE vision, information and new products. We also met some of their customers attendees to present our own product SafetyNex for Safe Navigation, Driver Assistance and Automated Driving.

Nexyad is a partner of HERE Technologies.

Paris, May 10th 2024.

Autonomous Driving boosted by AI
Generative artificial intelligence is coming to cockpits, but AI has been present in cars for many years. In particular to improve safety and driving aids. The 100% autonomous everyday car is not for now but the vehicle boosted by AI to become more and more autonomous is the present and this is the subject of the discussion with our guests.

Nicolas Léonetti / KPMG
Gérard Yahiaoui / Nexyad
Presented Aurélien Fleurot.

La conduite Autonome boostée par l’IA
L’intelligence artificielle générative arrive dans les cockpits mais cela fait de nombreuses années que l’IA est présente dans les voitures. En particulier pour améliorer la sécurité, les aides à la conduite. La voiture du quotidien 100% autonome, ce n’est pas pour tout de suite mais le véhicule boosté par l’IA pour devenir de plus en plus autonome, c’est le présent et c’est l’objet de la discussion avec nos invités.

Extract : « NEXYAD has developed the prudence metric for autonomous driving. This changes everything, autonomous vehicles drive like experienced humans. […] We work on road events instead of working on scenarios which are combinations of events. I’ll give you an example: presence of a truck, a bicycle and the approach of a roundabout. With these three events, we can construct many scenarios. At Nexyad, we assign a prudence function for each event in space and time (where they are at any moment) and combine them formally to fit any scenario. By doing so, we save a huge amount of time while being more safe. »

Paris, April 25th 2024.
Nexyad CEO Gerard Yahiaoui recorded april 25th an interview at MOBILITY TV.
It will be soon broadcast, stay tuned…

Linas -Montlhéry, April 15th 2024.

UTAC rented Nexyad Dreamotor1 for a week of road safety tests about anticipation to some diffilculties. In particular, was tested the Dreamotor’s skills of its preventive ACC system as part of PRISSMA Project.
Thanks to Alain Piperno, Céline Serbouh, Michael Dessaint, Vignan Srikanthan and all UTAC staff.

UTAC est centre d’essais officiel Euro NCAP en France et possède également une place unique en Europe grâce à ses laboratoires d’essais accrédités ISO 17025.

NEXYAD Automotive & Transportation Newsletter #43, April 1st, 2024

It’s Time for BYOD & SDV

Headlines :

– NEXYAD CEO Editorial

– NEXYAD signed a co-Innovation Partnership with Stellantis

– Demonstration of NEXYAD Dreamotor1 to Partners

– New videos of Use-Cases in Dreamotor1 using SafetyNex

– Colloque sur la sécurité Routière / Road Safety Conference

Go to the Newsletter

Paris, March 20th 2024.

« 3500 morts par an sur les routes depuis 10 ans : et maintenant on fait quoi ? »

La ligue de défense des conducteurs a organisé un colloque pour faire le point sur les mesures gouvernementales depuis plusieurs décennies en matière de sécurité routière.
Il ressort des interventions des trois tables rondes, que l’Etat Français a beaucoup misé sur la répression, que l’état des routes en France se dégrade depuis la décentralisation et la perte en compétence des équipes de l’équipement éclatées par département. Les nouvelles mobilités, bicyclettes et trotinettes partageant souvent une même infrastruture ont augmenté l’accidentologie des vulnérables peu protégés. Et les moyens financiers pour une bonne éducation des jeunes conducteurs semblent manquer.
Les points positifs étant l’amélioration de la sécurité active et passive des véhicules depuis plus d’une vingtaine d’années et la prise en charge plus rapide et plus efficace des blessés de la route par les services de santé et de traumatologie.

The French Drivers’ Defense League organized a conference to take stock of government measures over several decades in terms of road safety. Despite multiplication of speed cameras and lower speed limits, the number of fatalities is stable for 10 year around 3500 each year in France.
It emerges from the interventions of the three round tables that the French State has relied heavily on repression, that the state of the roads in France has deteriorated since decentralization and the loss of skills of the equipment teams split up by department. New forms of mobility, bicycles and scooters often sharing the same infrastructure, have increased the number of accidents among poorly protected vulnerable people. And the financial means for a good education of young drivers seem to be lacking.
The positive points being the improvement in active and passive safety of vehicles over more than twenty years and the faster and more effective treatment of road injuries by health and trauma services.

Tous les détails sur le programme de la demi-journée et les intervenants sont ici (all details here) : https://www.liguedesconducteurs.org/actu-1/colloque-3-500-morts-par-an-sur-les-routes-depuis-dix-ans

Guyancourt, March 7th 2024.

We had the pleasure to carry two staff member of Renault for a test ride of Dreamotor1 around TechnoCentre.

SafetyNex provides in real time prudent recommended speed, lack of prudence alerts and instant prudent speed for Driver Assistance. See an example of use-case below :

This video was made in our democar Dreamotor1.
On the screen we can see the HMI (for engineers) with the road captured by a dashcam. On the top left of the screen there are the current speed of the vehicle, the speed limit sign, and just below the next speed limit read on the map.
The colored square on the left says the driving is prudent, it can be yellow when it’s time to be vigilant and even red if driving is at risk. When a singularity is detected, the road sign is pasted with the distance where to apply recommended speed in order stay prudent.
Below the colored square, risk target indicates the level of max accepted risk to stay green. In this example, we vary the risk target to see adaptation of the system to different drivings : from very prudent to prudent.
On the top center, we show action of the blinkers indicating new direction at the Electronic Horizon.

Saint Germain en Laye, February 13th 2024.

Nexyad invited a Global Logic executive for a test with Dreamotor1 today, on open roads around our office in St Germain en Laye.

SafetyNex provides in real time prudent recommended speed, lack of prudence alerts and instant prudent speed for predictive ACC and Autonomous Vehicle. SafetyNex goes further than Intelligent Speed Assitance ISA (which is just a reminder of speed limit signs on roads), SafetyNex gives the prudent speed by anticipation everywhere it can be tricky for the driver and his/her passengers.