SLIDE 1
Autonomous Cars for City Traffic
Raúl Rojas and the AutoNOMOS Team
Freie Universität Berlin Fachbereich Mathematik und Informatik Dahlem Center for Intelligent Systems
Autonomous Cars for City Traffic Ral Rojas and the AutoNOMOS Team - - PowerPoint PPT Presentation
Autonomous Cars for City Traffic Ral Rojas and the AutoNOMOS Team - - PowerPoint PPT Presentation
Autonomous Cars for City Traffic Ral Rojas and the AutoNOMOS Team Freie Universitt Berlin Fachbereich Mathematik und Informatik Dahlem Center for Intelligent Systems Topics Motivation The experimental vehicles Sensors Navigation and
SLIDE 2
SLIDE 3
Topics
Motivation The experimental vehicles Sensors Navigation and control Autonomous cars are green cars
SLIDE 4
The past of transportation
The Great Horse-Manure Crisis In New York in 1900, the population of 100,000 horses produced 2.5 million pounds of horse manure per day
SLIDE 5
Automobile 1900
SLIDE 6
Green Cars: The End of Cheap Oil
SLIDE 7
The future of transportation
SLIDE 8
II The Vehicles
SLIDE 9
SLIDE 10
Team Berlin (FU Berlin)
SLIDE 11
Sensors in „Spirit of Berlin“
SLIDE 12
Sensors
SLIDE 13
MadeInGermany
SLIDE 14
MIG Sensors
SLIDE 15
15
Sensors in MadeInGermany
SLIDE 16
e-INSTEIN: electric & intelligent
SLIDE 17
Electronics in the trunk
SLIDE 18
iPad Remote Control
SLIDE 19
iPad Features
SLIDE 20
III Sensors
- GPS and IMU Navigation
- Laser scanners – two dimensional
- 3D Laser scanner
- Video Cameras
SLIDE 21
GPS Positioning
- IMU: Inertial Measurement Unit generates a true
representation of vehicle motion in all three axes
SLIDE 22
Differential GPS in Germany
SLIDE 23
Ibeo Laser Scanner
200 meter range
SLIDE 24
Velodyne
SLIDE 25
One important city feature: trees, poles
SLIDE 26
III Computer Vision
SLIDE 27
Interior / Front Cameras
SLIDE 28
Automatic Sensor Calibration
SLIDE 29
Object recognition
AdaBoost Stereo Traffic lights (Video)
SLIDE 30
Traffic Lights are Recognized
SLIDE 31
Control
- Simulator for vehicle dynamics
- High-level navigation planner
- Low-level reactive control
SLIDE 32
Traffic rules are followed
SLIDE 33
Pedestrians are Detected
SLIDE 34
Architecture
SLIDE 35
35
Route Network Definition File (RNDF)
- Street Segments
– GPS-Points – Lane width – Maximum speed
- Crossings
– Entry-Exit pairs – Stop signs – Traffic lights
SLIDE 36
36
AutoNOMOS – Macroplan (KN)
- Macroplan: BFS in
a graph
- Limited depth of
plan (150 m)
- Microplan for each
segment
SLIDE 37
37
AutoNOMOS – Microplan I
central shift / middle shift lateral shifts
maxShiftDist
nudges
swerveDist
swerves
swerveDist
SLIDE 38
Microplan - Transformation
38 38
Apr 11, 2012
Transformation
SLIDE 39
39
Microplan (KN)
- Next few meters
(50m)
- Attached to lane spline
- Evaluation function selects
SLIDE 40
40
AutoNOMOS – Evaluation function
Weighted sum of:
– non-collision [0,1] – Distance to checkpoint – time (t =s/v) – Curvature (1/r) – Maneuver value – heuristics
SLIDE 41
41
GUI – Controller/Behaviour
SLIDE 42
- Prof. Dr. Raúl Rojas, Freie Universität Berlin, Institut für Informatik - "AutoNOMOS"
42
Parking and maneuvering
SLIDE 43
Experiments in Traffic
SLIDE 44
SLIDE 45
Autonomous Wheelchair
2 x Lidar 2 x Odometer Kinect 2 x Computer (Linux & Windows) Ethernet CAN-Bus
SLIDE 46
Control
Eye tracking iPhone iPad Brain-Computer-Interface
SLIDE 47
Sensor data
Odometer (left & right): rotary encoders measure travelled distances . Laserscanner(front & rear): Distances (8cm above the ground) Kinect: RGB- and depth image.
SLIDE 48
SLIDE 49
SLIDE 50
SLIDE 51
SLIDE 52
Evolution vs Revolution: Driver Assistance Systems
bester-fahrer.de
SLIDE 53
- Prof. Dr. Raúl Rojas, Freie Universität Berlin, Institut für Informatik - "AutoNOMOS"
53