Road curvature estimation system
View Patent ↗A processor operatively coupled to a speed sensor adapted to generate a measure of a longitudinal speed of a vehicle on a roadway, and to a source of a measure of yaw rate of the vehicle, provides for selecting a most likely roadway model of a plurality of different roadway models and for outputting a corresponding associated at least one curvature parameter as an estimate of curvature of the roadway, wherein the processor incorporates a plurality of curvature estimators associated with the corresponding plurality of different roadway models.
1. A road curvature estimation system, comprising:
a. a speed sensor adapted to measure a longitudinal speed of a vehicle on a roadway and to generate a measure of longitudinal speed responsive thereto;
b. a source of a measure of yaw rate of said vehicle; and
c. a processor operatively coupled to said speed sensor and to said source of said measure of yaw rate, wherein said processor comprises a plurality of curvature estimators associated with a corresponding plurality of different roadway models, each said curvature estimator is adapted to estimate at least one associated curvature parameter of a corresponding said different roadway model, said processor provides for selecting a most likely of said different roadway models as a most likely roadway model, and said processor provides for outputting said at least one curvature parameter of said most likely roadway model as an estimate of curvature of said roadway.
2. A road curvature estimation system as recited in claim 1 , wherein said processor comprises an interacting multiple model algorithm incorporating a mode jump process characterized by a Markov chain.
3. A road curvature estimation system as recited in claim 2 , wherein said interacting multiple model algorithm comprises:
a. a mode-conditioned filter initialization process;
b. a mode-conditioned state propagation process;
c. a combination of mode-conditioned state and covariance propagations;
d. a mode-conditioned update process;
e. a probability evaluation and update process; and
f. process to output state and covariance estimates.
4. A road curvature estimation system as recited in claim 1 , wherein said plurality of different roadway models comprise at least two roadway models selected form a straight road model, a quadratic road model, and a clothoid road model.