Combined road prediction and target tracking in collision avoidance
read more
Citations
Statistical Threat Assessment for General Road Scenes Using Monte Carlo Sampling
Creating Enhanced Maps for Lane-Level Vehicle Navigation
Toward Autonomous Collision Avoidance by Steering
Estimation of Nonlinear Dynamic Systems : Theory and Applications
A novel multi-lane detection and tracking system
References
A Curvature-based Scheme for Improving Road Vehicle Guidance by Computer Vision
Robust car tracking using Kalman filtering and Bayesian templates
Advanced lane recognition-fusing vision and radar
Sensor fusion for improved vision based lane recognition and object tracking with range-finders
Related Papers (5)
Sensor fusion for improved vision based lane recognition and object tracking with range-finders
Robust car tracking using Kalman filtering and Bayesian templates
Frequently Asked Questions (8)
Q2. What is the maximum clothoid for the curves?
The recommended maximum clothoid parameters for these curves are given by the formulac1 = k v3where k = 0.45 (m/s3) which is the maximum ”jerk” and v the velocity, giving the clothoid 1.7 · 10−4 and 2.9 · 10−5 (1/m2) for the 50 km/h and the 90 km/h curve respectively.
Q3. how can the authors obtain a relationship between the two?
The authors can obtain a relationship between the two by taking the time derivative of ΨrelΨrel = Ψabs + γ ⇒ Ψ̇rel = Ψ̇abs + γ̇ = Ψ̇abs +v r = Ψ̇abs + c0vwhere r is the current road radius, v the velocity and γ denotes the angle between the lane and some fix reference.
Q4. what is the axis of the road?
In the longitudinal direction the authors will use the ẍi = 0− ahost cosΨrel , ahost being the measured acceleration of the host vehicle so that with sample timeTs the authors get the motion equations:xit+1 = x i t +Tsẋ i t +ahost,t cosΨrel,tT 2 s /2+w1,t ẋit+1 = ẋ i t +ahost,t cosΨrel,tTs +w2,t (7a) yit+1 = y i t +w3,tFor the road geometry parameters the authors first clarify that Ψrel is the angle offset to the lane and Ψabs is the angle to some fix reference.
Q5. How important is the quality of the lane geometry estimate?
This is where the quality of the lane geometry estimate becomes utterly important, even the slightest error in heading angle or curvature will result in a significant lateral error for vehicles at a long distance, say 70 - 100 meters.
Q6. What is the reason for the improvement of the performance of the Kalman filter?
This is intuitive, if bad visibility was detected by for example the vision system, you would typically increase the process noise of road measurements in the Kalman filter in order to rely more on other measurements and on the motion model.
Q7. What is the tangent vector for t?
Of course, when approaching a curve the authors might, for example, have situations were the section 0 - 50 meters of their field of view is a straight line and the section 50 - 100 meters is a clothoid, a case which can not be modelled with a linear curvature law.
Q8. What is the way to improve the performance of the Kalman filter?
It can be seen in the Figures 4 and 5 that the optimal performance is reached at higher measurement noise for the bad visibility case than the good visibility case.