DATA-DRIVEN CONTROL FOR AUTONOMOUS DRIVING
Techniques are described to determine parameters and/or values for a control model that can be used to operate an autonomous vehicle, such as an autonomous semi-trailer truck. For example, a method of obtaining a data-driven model for autonomous driving may include obtaining data associated with a first set of variables that characterize movements of an autonomous vehicle over time and commands provided to the autonomous vehicle over time, determining, using at least the first set of data, non-zero values and an associated second set of variables that describe a control model used to perform an autonomous driving operation of the autonomous vehicle, and calculating values for a feedback controller that describes a transfer function used to perform the autonomous driving operation of the autonomous vehicle driven on a road.
1 . A method of autonomous driving control model, comprising:
receiving a first set of data obtained from sensors located on an autonomous vehicle,
wherein the first set of data includes a plurality of driving related commands provided to the autonomous vehicle and a plurality of velocities of the autonomous vehicle, and
wherein each driving related command and each velocity is associated with a timestamp;
generating a first matrix comprising the plurality of driving related commands, the plurality of velocities, wherein each column of the first matrix is associated with a unique timestamp, wherein each row of the first matrix includes data associated with a same timestamp;
generating a second matrix comprising a plurality of acceleration determined from the plurality of velocities;
determining a third matrix by using a minimization function with the first matrix and the second matrix, wherein a set of non-zero values from the third matrix describe a control model; and
determining values for a controller that describes a transfer function used to perform a driving related operation of the autonomous vehicle,
wherein the values for the controller are based on the set of non-zero values of the control model, and
wherein the controller sends a first throttle command to an engine of the autonomous vehicle based on the transfer function and a set of values obtained from a sensor.
2 . The method of claim 1 , wherein the first matrix includes constant values and polynomial functions of the plurality of driving related commands and the plurality of velocities.
3 . The method of claim 1 , wherein the plurality of velocities include a first set of velocities related to one part of the autonomous vehicle, and a second set of velocities related to another part of the autonomous vehicle.
4 . The method of claim 3 , wherein the first set of velocities includes a yaw rate of a front region of the autonomous vehicle, and wherein the second set of velocities includes an angular velocity of a rear region of the autonomous vehicle.
5 . The method of claim 1 , wherein the plurality of velocities include a longitudinal velocities and lateral velocities.
6 . The method of claim 1 , wherein the plurality of driving related commands include a steering angle command and a second throttle command.
7 . The method of claim 1 ,
wherein the first set of data is sampled at a first rate in response to the autonomous vehicle operating in a changing environment, and
wherein the first set of data is sampled at a second rate that is lower than the first rate in response to the autonomous vehicle operating in a stationary environment.
8 . A non-transitory computer readable storage medium having code stored thereon, the code, when executed by a processor, causing the processor to implement a method comprising:
receiving a first set of data obtained from sensors located on an autonomous vehicle,
wherein the first set of data includes a plurality of driving related commands provided to the autonomous vehicle and a plurality of velocities of the autonomous vehicle, and
wherein each driving related command and each velocity is associated with a timestamp;
generating a first matrix comprising the plurality of driving related commands, the plurality of velocities, wherein each column of the first matrix is associated with a unique timestamp, wherein each row of the first matrix includes data associated with a same timestamp;
generating a second matrix comprising a plurality of acceleration determined from the plurality of velocities;
determining a third matrix by using a minimization function with the first matrix and the second matrix, wherein a set of non-zero values from the third matrix describe a control model; and
determining values for a controller that describes a transfer function used to perform a driving related operation of the autonomous vehicle,
wherein the values for the controller are based on the set of non-zero values of the control model, and
wherein the controller sends a first throttle command to an engine of the autonomous vehicle based on the transfer function and a set of values obtained from a sensor.
9 . The non-transitory computer readable storage medium of claim 8 , wherein the first matrix includes constant values and polynomial functions of the plurality of driving related commands and the plurality of velocities.
10 . The non-transitory computer readable storage medium of claim 8 , wherein the plurality of velocities include a first set of velocities related to one part of the autonomous vehicle, and a second set of velocities related to another part of the autonomous vehicle.
11 . The non-transitory computer readable storage medium of claim 10 , wherein the first set of velocities includes a yaw rate of a front region of the autonomous vehicle, and wherein the second set of velocities includes an angular velocity of a rear region of the autonomous vehicle.
12 . The non-transitory computer readable storage medium of claim 8 , wherein the plurality of velocities include a longitudinal velocities and lateral velocities.
13 . The non-transitory computer readable storage medium of claim 8 , wherein the plurality of driving related commands include a steering angle command and a second throttle command.
14 . The non-transitory computer readable storage medium of claim 8 ,
wherein the first set of data is sampled at a first rate in response to the autonomous vehicle operating in a changing environment, and
wherein the first set of data is sampled at a second rate that is lower than the first rate in response to the autonomous vehicle operating in a stationary environment.
15 . The non-transitory computer readable storage medium of claim 8 , wherein the third matrix is determined by using the minimization function by determining fewest parameters in equations governing vehicle dynamics for the control model by using a sparse regression technique.
16 . A system, comprising:
a processor, and
a memory comprising instructions that, when executed by the processor, configure the processor to:
receive a first set of data obtained from sensors located on an autonomous vehicle,
wherein the first set of data includes a plurality of driving related commands provided to the autonomous vehicle and a plurality of velocities of the autonomous vehicle, and
wherein each driving related command and each velocity is associated with a timestamp;
generate a first matrix comprising the plurality of driving related commands, the plurality of velocities, wherein each column of the first matrix is associated with a unique timestamp, wherein each row of the first matrix includes data associated with a same timestamp;
generate a second matrix comprising a plurality of acceleration determined from the plurality of velocities;
determine a third matrix by using a minimization function with the first matrix and the second matrix, wherein a set of non-zero values from the third matrix describe a control model; and
determine values for a controller that describes a transfer function used to perform a driving related operation of the autonomous vehicle,
wherein the values for the controller are based on the set of non-zero values of the control model, and
wherein the controller sends a first throttle command to an engine of the autonomous vehicle based on the transfer function and a set of values obtained from a sensor.
17 . The system of claim 16 , wherein the first matrix includes constant values and polynomial functions of the plurality of driving related commands and the plurality of velocities.
18 . The system of claim 16 , wherein the plurality of velocities include a first set of velocities related to one part of the autonomous vehicle, and a second set of velocities related to another part of the autonomous vehicle.
19 . The system of claim 18 , wherein the first set of velocities includes a yaw rate of a front region of the autonomous vehicle, and wherein the second set of velocities includes an angular velocity of a rear region of the autonomous vehicle.
20 . The system of claim 16 , wherein the plurality of velocities are measured by an inertial measurement unit (IMU) device on the autonomous vehicle.