Estimating the Driving State of Oncoming Vehicles From a Moving Platform Using Stereo Vision IEEE Intelligent Transportation Systems 2009 M.S. Student,
Post on 21-Dec-2015
216 Views
Preview:
Transcript
- Slide 1
- Estimating the Driving State of Oncoming Vehicles From a Moving Platform Using Stereo Vision IEEE Intelligent Transportation Systems 2009 M.S. Student, Heejong Hong 07. 14. 2014 Alexander Barth and Uwe Franke
- Slide 2
- Introduction Related Works Proposed Method Experimental Results Conclusion Outline 2
- Slide 3
- Introduction Driver-assistance and safety systems http://www.6d-vision.com/home/bedeutung Dynamic Object Detection for DAS Safety System with Dynamic Path Estimation
- Slide 4
- 1. A model-free object representation based on groups 2. Fusion active sensors 3. Track-before-detection 4. Rediscovering an image region labeled as vehicle Related Works 4 D. Beymer, P. McLauchlan, B. Coifman and J. Malik "A real-time computer vision system for measuring traffic parameters", Proc. Comput. Vis. Pattern Recog, pp.495 -501 1997 M. Maehlisch, W. Ritter and K. Dietmayer "De-cluttering with integrated probabilistic data association for multisensor multitarget ACC vehicle tracking", Proc. IEEE Intell. Veh. Symp., pp.178 -183 2007 U. Franke, C. Rabe, H. Badino and S. Gehrig "6D-vision: Fusion of stereo and motion for robust environment perception", Proc. 27th DAGM Symp., pp.216 - 223 2005 X. Li, X. Yao, Y. Murphey, R. Karlsen and G. Gerhart "A real-time vehicle detection and tracking system in outdoor traffic scenes", Proc. 17th Int. Conf. Pattern Recog., pp.II:761 -II:764 2004 1.2.3.
- Slide 5
- 5 Proposed Method
- Slide 6
- Object Model 1. Pose (relative orientation and translation to ego-vehicle) 2. Motion State (velocity, acceleration, yaw rate) 3. Shape (rigid 3-D point cloud) Motion State Shape Pose
- Slide 7
- Object Tracking Extended Kalman Filter (EKF) : Kalman filter for nonlinear model State transition(f) and observation model(h) Discrete-time predict and update equations Wikipedia : http://en.wikipedia.org/wiki/Extended_Kalman_filter Jacobian of system & measurement model Example)
- Slide 8
- Object Tracking 1. State Vector of an object instance Reference point in ego-coordinates Rotation point in object-coordinates The object origin is ideally defined on the center rear axis
- Slide 9
- Object Tracking 2. Dynamic(System) Model R is 3x3 rotation matrix around the height axis N. Kaempchen, K. Weiss, M. Schaefer and K. Dietmayer "IMM object tracking for high dynamic driving maneuvers", Proc. IEEE Intell. Veh. Symp., pp.825 -830 2004 Predicted state vector Time-discrete system EquationTransformation of an object point Translation matrix
- Slide 10
- Object Tracking 3. Measurement Model The measurement nonlinear eq. : perspective camera model Jacobian of measurement model Objects feature points on image coordinates using feature tracker (KLT) Feature point tracking using KLT
- Slide 11
- Kalman Filter Initialization 1. Image Based Initialization 2. Radar-Based Initialization (detect oncoming vehicle up to 200m) The mean velocity vector : The centroid of the 3-D positions : Initial Yaw : The lateral and longitudinal positions of the radar target :, Absolute radar velocity of the object : Initial Yaw :
- Slide 12
- Point Model Update 1. Maximum-likelihood estimation 2. Simple average filter Objects Shape Expectation = 3x3 covariance matrix of t Expected objects shape
- Slide 13
- 13 Experimental Result
- Slide 14
- Simulation Results Synthetic Sequence
- Slide 15
- Real World Results Country Road Curve I
- Slide 16
- Real World Results Country Road Curve II
- Slide 17
- Real World Results Oncoming Traffic at Intersections
- Slide 18
- Real World Results Leading Vehicles & Partial Occlusions
- Slide 19
- Real World Results Challenges and Limits
- Slide 20
- Joint Histogram Based Cost Aggregation for Stereo Matching - TPAMI 2013 20 Conclusion
- Slide 21
- Joint Histogram Based Cost Aggregation for Stereo Matching - TPAMI 2013 21 Contribution New method for the image-based real-time tracking (25Hz, 640x480) Results of experiments with synthetic data & real-world Two different object detection method (image & radar) Feature-based object point model does not require a priori knowledge about the objects shape Weakness No specific system block diagram User defined rotation point Shape depends on outlier removing algorithm (ex : max distance parameter) Shape is very sensitive about outlier of point cloud (because of yaw)
- Slide 22
- 22 Thank you!
top related