Nexyad est fier d’annoncer que l’Institut VeDeCoM, Véhicule Décarboné et Communiquant, vient d’acquérir des licences utilisateurs des logiciels RoadNex et ObstaNex.
L’institut VeDecom monte en puissance depuis sa création et entend jouer un rôle majeur dans les domaines des véhicules autonomes, connectés et électriques. Sa stratégie d’acquisition de l’état de l’art en vision et intelligence artificielle pour les ADAS profite à Nexyad au travers de ses produits de détection de route et détection d’obstacles.
Among the few companies that worked on ADAS and Autonomous Vehicle since a long tima, NEXYAD is one that have the most cumulated experience.
1993 : scientific paper by Gerard Yahiaoui (Founder of NEXYAD with Pierre Da Silva Dias) with a researcher of the Automotive Company PSA Peugeot Citroën : « Texture-based Image Segmentation for Road Recognition with Neural Networks », G. Yahiaoui, M. de Saint Blancard, Sixth international conference on neural networks and their industrial & cognitive applications NeuroNîmes93, EC2, Nîmes, 1993,
2007 : participation of NEXYAD to the DARPA Challenge (Autonomous Vehicle) in the team « Blue Froggy » (with INRIA and INDUCT).
2016 : road detection, obstacle detection, road safety estimation, visibility measurement, available on smartphones (in March).
Experience matters !
What your company was doing in 1993 in the field of ADAS ?
What your company was doing in 2007 about Autonomous Cars ?
NEXYAD accumulated work on artificial vision based ADAS since more than 20 years, even when nobody believed that one day it will be camera inside cars.
METHODOLOGY for ADAS VALIDATION :
Potential Contribution of Other Scientific Fields Which Have Already Answered the Same Questions
By Gérard YAHIAOUI & Pierre DA SILVA DIAS
This volume by Jochen Langheim collects selected papers of the 3rd CESA Automotive Electronics Congress, Paris 2014. CESA is the most important automotive electronics conference in France. The topical focus lies on state-of-the-art automotive electronics with respect to energy consumption and autonomous driving. The target audience primarily comprises industry leaders and research experts in the automotive industry.
Les Compagnies d’Assurances, les Constructeurs Automobiles et leur équipementiers ont tous des raisons pour mesurer le comportement du conducteur. Cela a déjà commencé, avec les boîtiers des assureurs dans le cadre du fameux « pay how you drive » : l’idée des compagnies d’assurance automobile est de moduler les tarifs en fonction de la dangerosité des conduites : un boîter embarqué dans le véhicule (ou une App sur le smartphone du conducteur) note le style de conduite (en utilisant les accéléromètres intégrés dans l’électronique), et lui attribue un degré de risque d’accidents.
Or, le style de conduite (exemple : freinages sévères, ou au contraire conduite coulée) n’est pas corrélé au risque d’accidents. Cela a été démontré formellement au travers de programmes de recherche collaborative (en particulier dans le cadre du programme national PREDIT sur la Sécurité Routière), et cela correspond aussi au « bon sens » : quelqu’un qui grille un stop sans faire de freinage sévère, sans accélérer comme un fou, tranquillement, à 40 km/h, est extrêmement dangereux et son comportement de conduite est fortement accidentogène, alors qu’il conduit « calmement ».
De même quelqu’un qui ne tient absolument pas compte des passages piétons (y compris lorsqu’un piéton est engagé) et le passent tranquillement sans freiner on un comportement, d’une part interdit par le code de la route, et d’autre part très accidentogène (avec des accidents mortels car il s’agit d’accident « véhicule vs vulnérable »).
Tous les boîtiers automobiles actuels qui utilisent le style de conduite comme clé de dangerosité sont donc absolument inefficaces et remontent une information fausse.
Tant qu’on ne rapproche pas le style de conduite du contexte géométrique, géographique, météo, état du conducteur etc …, on ne peux absolument rien conclure d’un style de conduite, à part pour la partie dite « Eco » (plus on conduit brutalement, moins on conserve l’inertie du véhicule, plus on consomme de carburant, et donc plus on rejette aussi de CO2).
NEXYAD always looking at surveys about ADAS and driverless cars market … Here is a relevant study about Automotive Advanced Driver Assistance Systems Market (2015-2025), by ReportBuyer.com
The Next Step Towards Autonomous, Self-Driving & Driverless Cars
LONDON, May 12, 2015 /PRNewswire/ — Report Details
In many ways, advanced driver assistance systems (ADAS) are the first step on the road to semi-autonomous or fully-autonomous self-driving and driverless vehicles. In 2015, Park Assistance, Surround-View Cameras, and Adaptive Cruise Control applications dominate the ADAS market; however, the penetration into passenger cars is relatively low despite the increasing rate of installations into the premium car segment manufacturers.
Often these are fitted in isolation, and even where multiple systems are fitted in a vehicle, great care is taken to ensure that these systems are perceived only as assisting the driver (who must remain fully alert) rather than taking over his role. As the number and abilities of these systems increases, the situation will move further along a continuum from driver assistance to in effect driver substitution, meaning that the driver would no longer be required to be alert and instead the onus would fall on the vehicle to warn the driver when his input was required, bringing him back to a state of full alertness.
ROBUSTNESS OF NEXYAD SOFTWARE MODULE FOR ROAD DETECTION : RoadNex
By NEXYAD
Detection of the road, detection of the lane, in front of the vehicle is now a « must-have »
for Advanced Driver Assistance Systems (ADAS) and of course for Autonomous Cars too.
Every R&D team is able to show cases of good detection. The difference between different
modules is robustness : ability to work in many cases (almost every cases).
For instance, robustness consideration led many big Automotive firms to interger the MOBILEYE
detection system : jus because MOBILEYE is more robust than detection systems developed by
those big firms. And robustness is not a matter of deployment : you won’t get a more robust
module is you put 10 000 developers on the project. You need time, big amount of data, and
« smart ideas ».
Note : This robustness definition leads to question on ADAS validation (« almost » every case is
not that well defined … how could we put some maths on those words). NEXYAD has been
developing an applied maths-based methodology for ADAS validation and is currently
recording a validation data base that will be soon available for free worldwide on the internet.
But let’s go back to road detection modules comparison.
There is another difference between road detection systems : do they need white markings
or are they able to work even without markings ?
NEXYAD founders has been working on road detection since the beginning of the 90’s and never
stopped (*). The NEXYAD team is one of the moste experienced team in the world about road detection.
That actually makes the difference, and RoadNex is a module that would take long to develop by
other teams. RoadNex is currently available on PC (windows, Linux) in the real time framework
RT-MAPS. RoadNex will be soon available :
. on electronic device of an Automotive Tier One Company
. on smartphones (so it works in real time on a smartphone usual processor ! try to compare to other modules)
(*) publication at a scientific congress in France in 1993 :
« Texture-based Image Segmentation for Road Recognition with Neural Networks », G. Yahiaoui, M. de Saint Blancard,
Sixth international conference on neural networks and their industrial & cognitive applications NeuroNîmes93, EC2,
Nîmes, 1993,
In order to have an idea of what robustness means, here are some case used to test RoadNex :
How many kilometers should you drive to sample those few road scenes variations ?
Autonomous Vehicle TEST & DEVELOPMENT Symposium 2016
31 may – 2 june 2016 Stuttgart, Germany.
2016 Preliminary Conference Programme
Wesnesday 1st June
09:15 – 15:45 – Test and Validation Strategies for Autonomous Vehicles
Room B
09:45 – Building a relevant validation database for camera-based ADAS
Gérard Yahiaoui, President and CEO, Nexyad, France
Validation of camera-based artificial vision systems applied on open world is a very complex issue. An HD colour camera may generate more than 65 000 power 2 000 000 different images (information theory), so it is not possible to test every possible message. We propose a deterministic approach for building a validation database using the AGENDA methodology that was developed and published in the 1990s for neural network database (learn and test) design.
Some drivers are quiet but still dangerous (in a road safety point of view) :
Most people think that brutal drivers (strong braking, etc …) are dangerous. The hypothesis is that they do not anticipate enough and then they are
led to brake very strongly. Many Insurance Companies developped their own onboard « dirving style » estimation that scores this kind of hypothesis.
Nexyad RoadNex v2.2 – Road Detection on little road in Forest
RoadNex detecting the lane on the road with or without markings.
The green Arrow shows the possible direction to follow.
Nexyad provides modules for ADAS (Advanced Driver Assistance Systems) : some of those modules such as RoadNex road detection or ObstaNex obstacles detection are competitors of the famous modules of the company Mobileye.
Gérard Yahiaoui will present for NEXYAD a paper about Validation of ADAS at the next Autonomous Vehicle Test & Development Symposium 31 may – 2 June 2016 in Stuttgart, Germany.
Save the dates and come to visit the Nexyad booth at the Stuttgart Messe. http://www.autonomousvehiclesymposium.com/
RoadNex v2.2 Road Detection – Scooter passing by in City Traffic
RoadNex v2.2 Road Detection – Scooter passing by in City Traffic
RoadNex detecting the lane on the road with or without markings.
Here, see the red line that reduces with scooters and cars; and the red surface highlighted which stops with scooters and cars, telling us it’s not road anymore, so detecting obstacles in negative.
The green Arrow shows the possible direction to follow.
Nexyad provides modules for ADAS (Advanced Driver Assistance Systems) : some of those modules such as RoadNex road detection or ObstaNex obstacles detection are competitors of the famous modules of the company Mobileye.
NEXYAD RoadNex v2.2 Road Detection for ADAS : Green Desert Track
RoadNex detects the lane and the road surface and does not depend on the level of equipment of road infrastructure. Example here, on a dirt track in the middle of nowhere…
RoadNex v2.2 Road Detection for ADAS : Road without Marking
Nexyad RoadNex v2.2 for ADAS Road without Marking
RoadNex detects the lane and the road surface and does not depend on the level of equipment of road infrastructure.
Nexyad Automotive & Transportation has launched the blog NEXYAD-ADAS Solutions.
As the name says, it’s all about Advanced Driver Assistance Systems solutions.
SafetyNex by NEXYAD and EcoGyzer by NOMADIC SOLUTIONS together to estimate the Driver Behaviour Signature.
We ask an experimented driver to drive the same path four times with a different behaviour.
Here we show the Sportive-Safe Driver :
The Sportive-Safe Driver is not Eco, but he is safe most of the time. Despite, he accelerates strongly, he slows down when he arrives to intersections and pedestrian pathways. He has only four points in the red area of the signature mapping to compare with the Quiet-Dangerous Driver seen in an earlier news
SafetyNex & EcoGyzer : the Driver Behaviour Signature
SafetyNex is a Nexyad Module for ADAS that estimates the risk (or safety) in driving 100% correlated with accidentology. It is unique and disruptive in the whole world of Road Safety.
Recently, EcoGyzer a tool by the french company Nomadic Solutions, has been integrated to SafetyNex to estimate the Eco driving.
Now we are proud to announce that SafetyNex & EcoGyzer as a world premiere App estimating the Driver Behaviour Signature. To figure the Driver Behavour Signature, we imagined a special mapping as you’ll see on the demo films.
Every second in driving, one point is plotted on the mapping. The more right a point is, the more Eco is the driving; the more up, the more safe it is. At the end of the path, there is a cloud of points called the Driver Behaviour Signature.
In the following demo films, we show four different driving behaviour signatures for the same path :
The Good Driver
The Bad Driver
The Quiet-Dangerous Driver
The Sportive-Safe Driver
Comparison of the four Driver Behaviour Signatures
* * * * *
Nexyad ADAS Validation Database still in Progress
NEXYAD has been starting the development of a data base for artificial vision-based ADAS test and validation.
This data base will be relevant and unique because it is fully decribed in two ways :
. reality : position of road and obstacles
. driving situation (i.e. curve in a foggy weather with pedestrian crossing, …) using the methology AGENDA.
To read more :
– « Methodology for ADAS validation: Potential Contribution of other Scientific Fields which have already answered the Same Questions »,
G. Yahiaoui, P. Da Silva Dias, proceedings of the 3rd CESA Automotive Electronics Congress May 2014 Paris, Lecture Notes in Mobility,
ENERGY CONSUMPTION AND AUTONOMOUS DRIVING, Jochen Langheim Ed, Springer, pp 133-138.
– « Validation of Advanced Driving Assistance Systems », G. Yahiaoui, N. du Lac, SafetyWeek congress, Aschaffenburg, May 2015.
Business details :
A part of this data base will be soon available for free on the internet, to the worldwide ADAS and Autonomous
vehicle community (labos and firms).
If you wish to receive the link as soon as it is available, click HERE and fill the registration form.
(Free access to the NEXYAD Artificial Vision-based ADAS Validation Database)
The complete data base should be available soon through an annual membership.
For more information : contact NEXYAD Olivier BENEL +33 139 04 13 60
* * * * *
RoadNex v2.2 Robustness
On a new release v2.2 of RoadNex, we’ve added an arrow that shows the direction to be followed by the vehicle according to our road detection always defined by the edges of the rollable track and the surface of the rollable track. New tests were performed on videos of different types of roads, including tracks in the desert or lit roads by car lights at night.
These examples demonstrate the robustness of our module software that is more and more a reel competitor of lanes Detection System developed by Mobileye.
https://nexyad.net/Automotive-Transportation/wp-content/uploads/2015/10/Article-CCFA.png
To read more :
https://nexyad.net/Automotive-Transportation/wp-content/uploads/2015/10/Article-Cleo.png
We use cookies to understand how you use our site and to improve your experience. By continuing to use our site, you accept our use of cookies. AcceptPrivacy & Cookies Policy Read More
Privacy & Cookies Policy
Privacy Overview
This website uses cookies to improve your experience while you navigate through the website. Out of these, the cookies that are categorized as necessary are stored on your browser as they are essential for the working of basic functionalities of the website. We also use third-party cookies that help us analyze and understand how you use this website. These cookies will be stored in your browser only with your consent. You also have the option to opt-out of these cookies. But opting out of some of these cookies may affect your browsing experience.
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.
Any cookies that may not be particularly necessary for the website to function and is used specifically to collect user personal data via analytics, ads, other embedded contents are termed as non-necessary cookies. It is mandatory to procure user consent prior to running these cookies on your website.
