System and method for low speed lateral control of a vehicle
A method for providing low speed lateral steering control for a vehicle is disclosed. The method includes receiving sensor data corresponding to a road wheel angle, determining a planned vehicle path of travel, defining a road wheel angle search range based on a maximum road wheel angle rate, determining a steering control goal using the road wheel angle that tracks and measures a difference between a current vehicle path and the planned vehicle path, determining an optimal steering control signal using the road wheel angle and the steering control goal and providing the control signal to a steering controller.
1. A method for providing low speed lateral steering control for an autonomously driven or semi-autonomously driven vehicle, the method comprising:
receiving, from at least one vehicle sensor, sensor data corresponding to a road wheel angle;
determining, by one or more data processors, a plurality of waypoints indicating a planned vehicle path of travel;
defining, by the one or more data processors, a road wheel angle search range based at least in part on a maximum road wheel angle rate;
determining, by the one or more data processors, a steering control goal using the road wheel angle that tracks and measures a difference between a current vehicle path and the planned vehicle path and defining a relationship between a steering command and a cost function, the cost function including a cost associated with a radial error and a cost associated with a heading error;
determining, by the one or more data processors, an optimal steering control signal using the road wheel angle and the steering control goal; and
providing the control signal to a steering controller.
2. The method of claim 1 , wherein determining the optimal steering control signal includes using one-dimensional nonlinear optimization to minimize a path tracking error between the current vehicle path and the planned vehicle path.
3. The method of claim 2 , wherein minimizing the path tracking error includes minimizing both a heading error and a radial error.
4. The method of claim 1 , wherein the cost function is defined by the equation:
J=J L +J φ
where J L is the cost associated with the radial error and J φ is the cost associated with the heading error and J L and J φ are expressed with the following equations:
J
L
=
1
N
∑
j
=
1
N
w
j
L
Δ
L
i
2
J
φ
=
1
N
∑
j
=
1
N
w
j
L
Δφ
i
2
where w j is a tunable weighting factor that is selected based on an importance of a segment of the projected vehicle path and N is a number of waypoints.
5. An automotive vehicle, comprising:
a steering system;
at least one vehicle sensor;
a controller electrically connected to the steering system and the at least one vehicle sensor, the controller configured to
receive, from the at least one vehicle sensor, sensor data corresponding to a road wheel angle;
determine a plurality of waypoints indicating a planned vehicle path of travel;
define a road wheel angle search range based at least in part on a maximum road wheel angle rate;
determine a steering control goal using the road wheel angle that tracks and measures a difference between a current vehicle path and the planned vehicle path and defining a relationship between a steering command and a cost function, the cost function including a cost associated with a radial error and a cost associated with a heading error;
determine an optimal steering control signal using the road wheel angle and the steering control goal; and
provide the control signal to the steering system.
6. The automotive vehicle of claim 5 , wherein determining the optimal steering control signal includes using one-dimensional nonlinear optimization to minimize a path tracking error between the current vehicle path and the planned vehicle path.
7. The automotive vehicle of claim 6 , wherein minimizing the path tracking error includes minimizing both a heading error and a radial error.
8. The automotive vehicle of claim 5 , wherein the cost function is defined by the equation:
J=J L +J φ
where J L is the cost associated with the radial error and J φ is the cost associated with the heading error and J L and J φ are expressed with the following equations:
J
L
=
1
N
∑
j
=
1
N
w
j
L
Δ
L
i
2
J
φ
=
1
N
∑
j
=
1
N
w
j
L
Δφ
i
2
where w j is a tunable weighting factor that is selected based on an importance of a segment of the projected vehicle path and N is a number of waypoints.
9. A method for providing low speed lateral control to an automotive vehicle, the method comprising:
receiving, from a vehicle sensor, sensor data corresponding to a road wheel angle;
determining, by one or more data processors, a plurality of waypoints indicating a planned vehicle path of travel;
defining, by the one or more data processors, a road wheel angle search range based on a maximum road wheel angle rate of change;
designing a steering control goal using a cost function that reduces a difference between a current vehicle path and the planned vehicle path, wherein the cost function includes a cost associated with a radial error and a cost associated with a heading error, and wherein the cost function can be expressed in a nonsingular form as a function of the road wheel angle, one or more vehicle geometric parameters, and the plurality of waypoints;
determining an optimal steering control signal by minimizing the cost function subject to a constraint driven by the maximum road wheel angle rate; and
providing the control signal to a steering controller.
10. The method of claim 9 , wherein the cost function is defined by the equation:
J=J L +J φ
where J L is the cost associated with the radial error and J φ is the cost associated with the heading error and J L and J φ are expressed with the following equations:
J
L
=
1
N
∑
j
=
1
N
w
j
L
Δ
L
i
2
J
φ
=
1
N
∑
j
=
1
N
w
j
L
Δφ
i
2
where w j is a tunable weighting factor that is selected based on an importance of a segment of the projected vehicle path, L, and N is a number of waypoints.
11. The method of claim 9 , wherein determining the optimal steering control signal includes using one-dimensional nonlinear optimization to minimize a path tracking error between the current vehicle path and the planned vehicle path.
12. The method of claim 11 , wherein minimizing the path tracking error includes minimizing both the heading error and the radial error.
13. The method of claim 10 , wherein the cost function has a resolvable singularity at z≡tan δ=0.
14. The method of claim 13 , wherein J L and J φ are expressed with the following equations:
J
L
=
1
N
∑
j
=
1
N
w
j
L
[
(
x
j
2
+
2
x
j
b
+
y
j
2
)
tan
δ
-
2
y
j
L
[
(
x
j
+
b
)
2
+
y
j
2
]
tan
2
δ
-
2
y
j
L
tan
δ
+
L
2
+
L
2
+
b
2
tan
2
δ
]
2
and
J
φ
=
1
N
∑
j
=
1
N
w
j
φ
[
φ
j
+
a
tan
(
x
j
+
b
)
z
y
j
z
-
L
]
2
.