Internationaler Automobil Kongress 17.10.2016 Juergen Schweiger - Continental Intelligent Transportation Systems www.continental-automotive.com Division Interior
Mobility Yesterday and Tomorrow Source:??? Source: Autovia.com Source: Jboard.tv 2
Mobility Today Chongqing Handy-walk Augsburg Traffic lights for mobile phone using 1 In 2013, 18 percent of U.S. vehicle crashes with injuries were because of distracted driving 2 1 Picture: Thomas Hosemann/Stadtwerke Augsburg 2 Source: https://www.cdc.gov/motorvehiclesafety/distracted_driving/ 3
Change in Mobility Our vision is that by 2020 no one should be killed or seriously injured in a new Volvo car. Håkan Samuelsson, President and CEO, Volvo Cars Hyundai, Kia to triple range of green cars by 2020 Reuters 1 Million Green Cars on South Korean Roads By 2020 S. Korean energy ministry By 2021, the fleet average to be achieved by all new cars is 95 grams of CO 2 per kilometer. EU legislation Ford CEO expects fully autonomous cars by 2020 Mark Fields, CEO of Ford Five or six years from now we will be able to achieve true autonomous driving where you could literally get in the car, go to sleep and wake up at your destination. Elon Musk, CEO of Tesla Nissan to provide fully autonomous vehicles by 2020 Andy Palmer, Executive Vice President of California-based Nissan Motors Ltd. IAC 2016 21 October 2016
Dynamic ehorizon provide value to safety, eco, comfort on the way towards Autonomous Driving ehorizon client BE Interface Aggregation and enrichment of dynamic road event data from multiple sources 4sec Dynamic Event Service Map tile delivery and road data processing Access to cloud via API ehorizon data delivery Aggregation and enrichment of map data from multiple sources including incremental map updates HD Map Service Road Database Service Update and enrich map information Update and enrich road event information Crowd-sourced creation of road data and landmarks from probe vehicles including continuous near real-time updates 5
Tomorrow s Situation: Sensors, Maps and Online Data Dynamic ehorizon: The Electronic Co-Pilot 3. Extension of limited in-vehicle resources 1. Highly accurate map model provided and updated via the Backend 2. Extended preview information Close preview: 10 minutes 4. Fleet based data collection Car sensor range 100 300 m IAC 2016 21 October 2016 6
Das Bild kann nicht angezeigt werden. Dieser Computer verfügt möglicherweise über zu wenig Arbeitsspeicher, um das Bild zu öffnen, oder das Bild ist beschädigt. Starten Sie den Computer neu, und öffnen Sie dann erneut die Datei. Wenn weiterhin das rote x angezeigt wird, müssen Sie das Bild möglicherweise löschen und dann erneut einfügen. HD Map Service Aggregating data of map providers and frequent tile based map updates Service overview Map Suppliers HD Map Service Vehicles ehorizon client OEM Backend Map data ehorizon data delivery Global map data API Service features HD map data aggregation from different 3rd party map suppliers Handling of different HD map formats and generation of one map format (e.g. OEM specific, NDS or AMF) Continuous inclusion of updates of HD map data as provided by 3rd party map suppliers Delivery of HD map tiles into the vehicle to support dynamic ehorizon use -cases Full HD map data access for OEM through CoBe API IAC 2016 21 October 2016 7
Das Bild kann nicht angezeigt werden. Dieser Computer verfügt möglicherweise über zu wenig Arbeitsspeicher, um das Bild zu öffnen, oder das Bild ist beschädigt. Starten Sie den Computer neu, und öffnen Sie dann erneut die Datei. Wenn weiterhin das rote x angezeigt wird, müssen Sie das Bild möglicherweise löschen und dann erneut einfügen. Road Database Service 10cm accuracy vehicle localization and permanently fresh HD road data Service overview Vehicle sensors Road Database data Road Database Service Road Database data delivery Vehicle ehorizon client Service features Highly accurate vehicle positioning using our landmark based optical localization technology Permanently fresh HD road data service based on our crowd-sourcing mapping technologies Processed probe data collection from vehicles Consolidation of multi-sensor based imagery data continuously update the Road Database of the empty road network Create and update Road Database data by software without human labor Provide access to incremental HD road data updates for map suppliers and OEMs API OEM or Map Supplier Backend Road Database data delivery IAC 2016 8
Model Generation and Distribution One Complete Cycle (Illustrative) Road Furniture: Sensor Data Capturin g Sensor Data Processing & Feature Extraction RoadDB Modelling & Snippet Generation Snippet Purging, Alignment & Statistical Evaluation Speed limit signs Stop signs Yield signs Passing/no passing zone signs Traffic lights (the object, not a prediction of the current color) Variable traffic signs (the object, not a prediction of the current status) Etc. Road Geometry: RoadDB Data Delivery RoadDB Data Update Distribution RoadDB Model Generation 3D ribbon of the road in 3D space including sloping and cambering Road edges Lane boundaries for roads with paint Permeability of lane boundaries Barriers Averaged trajectories for passenger vehicles and trucks on all roads Virtual lanes of roads without paint and through intersections Beginning and ending / merging of painted and virtual lanes Stop lines Pedestrian crossings Etc. 9
Road Database Some use-cases Localisation (also bad weather) Maintaining and updating traditional maps Three steps of localization: 1. Search landmarks in real-world using expected location from road database 2. Match with landmarks in model 3. Compute coordinates in model 1 X X 2 X X X X 3 Road Database from Continental change detection enriched road f urniture additional geo-coverage alternative data source automatic maintenance HD map from any supplier 10
Road Database On the Street A-Sample Autobox 3-D visualization 2-D visualization In vehicle Installed A-sample hardware Visualization of Road Database processes current processing status of the A-sample connection status to the Server 2-D visualization of the database 3-D view that shows the car driving in its lane Server In-vehicle visualization Database and the car s real time GPS location from the backend Real time 3-D view of the car SVC210 Surround View Camera Real time 3-D view of the car 11
Dynamic Event Service Aggregation of crowd sourced vehicle sensor data and 3rd party content Vehicle sensors Dynamic Event Service ehorizon data delivery Vehicles ehorizon client OEM Backend Event data
Continental BU ITS Mission Statement Road data Hazard data Construction sites Traffic signs Traffic info Traffic flow Traffic control Traffic lights We bring your world to you safely while you are on the move, connecting you to the information you need, when you need it Intersections Parking Fuel/power Energy consumption stations Payment Guiding Delivery Reservation Fuelling Charging Washing Last Mile Services Car Sharing Fully AutomatedParking IAC 2016 13
The New Mobility Business Models Self-driving Function Individual Mobility Focus: Attractive Car Focus: Perfect Service 14
Digitalization Services will become part of our product line Mistakes to avoid Previous Innovators Misread How Their Industry was Evolving Functionality that customers don t want to pay for. Underestimating security and privacy risks. Failing to anticipate new competitive threats. Waiting too long to get started. Overestimating internal capabilities. Changes will challenge Our Technologies Internet of Things Our Skill sets, mindsets, tools and processes Our customers will change Our organizational structure Partnering and M&A will be mandatory 15
Jürgen Schweiger Continental Automotive GmbH Intelligent Transportation Systems Siemensstrasse 12, D-93055 Regensburg, Germany Phone: +49 160 90 94 00 59 juergen.schweiger@continental-corporation.com 16