Methods, systems, and apparatuses for behavioral based adaptive cruise control (ACC) to driver's vehicle operation style

1. A method for implementing adaptive cruise control (ACC) established by Reinforcement Learning (RL) comprising:

executing, by a processor, adaptive cruise control to receive a set of vehicle inputs about a host vehicle's operating environment and current operations;

identifying, by the processor, a target vehicle operating in the host vehicle environment and quantifying a set of target vehicle parameters about the target vehicle derived from sensed inputs;

modeling, by the processor, a state estimation of the host vehicle and the target vehicle by generating a set of speed and torque calculations about each vehicle, and by utilizing a first time vector of ACC torque and a second time vector of detection of a brake light of the target vehicle;

generating, by the processor, a set of results from at least one reward function based on one or more modeled state estimations of the host vehicle and the target vehicle; and

processing the set of results with driver behavior data established by the RL to correlate one or more control actions to the driver behavior data;

wherein the generating of the set of results is performed from a plurality of reward functions comprising:

a first reward function R 1 pertaining to driver feedback and based on a difference between a driver torque (τ Driver ) and an ACC torque (τ ACC );

a second reward function R 2 pertaining to an indication and based on the first time vector of ACC torque ( τ Acc ) and the second time vector of detection of the brake light of the target vehicle ( b );

a third reward function R 3 pertaining to distance and based on a distance (d) between the host vehicle and the target vehicle; and

a fourth reward function R 4 pertaining ACC performance and based on an ACC speed ((v ACC ) and an estimated speed (v est ).

2. The method of claim 1 , further comprising:

applying, by the processor, at least one control action that is correlated to the driver behavior data established by RL to adjust at least one operation of the adaptive cruise control of the host vehicle.

3. The method of claim 2 , further comprising:

adjusting, by the processor, the at least one control action that is correlated to the driver behavior data established by RL based on a control safety check.

4. The method of claim 3 , further comprising:

updating, by the processor, data of a learning matrix based on the generated set of results from the at least one reward function to create a profile of driver behavior.

5. The method of claim 4 , further comprising:

calculating, by the processor, the reward function using a set of parameters comprising speed and torque calculations of estimations in acceleration and speed of the host and target vehicles.

6. The method of claim 5 , further comprising:

adjusting, by the processor, one or more distances between the host vehicle and the target vehicle based on learned driver behavior contained in the data of the learning matrix.

7. The method of claim 6 , wherein the control safety check comprises a difference in speed between a safe speed and an estimation of the speed of the target and host vehicles.

8. The method of claim 1 , wherein the modeling of the state estimation is further performed based on an identification of road geometries.

9. The method of claim 1 , wherein:

the first reward function R1 is calculated in accordance with the following equation R1=(τ ACC −τ Driver ) −2 ;

the second reward function R2 is calculated in accordance with the following equation R 2 =Convolution ( τ Acc * b );

the third reward function R3 is calculated in accordance with the following equation R 3 =d 2 ; and

the fourth reward function R4 is calculated in accordance with the following equation R 4 =(v ACC −v est ) −2 .

10. A system comprising:

a set of inputs obtained by a processor that comprise a set of vehicle inputs of one or more measurement inputs of a host vehicle operation and sensed inputs of an operating environment about a host vehicle used to execute control operations of a vehicle adaptive cruise control (ACC) system established by Reinforcement Learning (RL) and contained in the host vehicle;

the vehicle ACC system instructed by a driver behavior prediction model implemented by the processor established by RL that learns the driver's expectation online, and uses a neural network (NN) to process the set of vehicle inputs to adjust control operations;

the processor configured to identify a target vehicle operating in the host vehicle environment to quantify a set of target vehicle parameters about the target vehicle derived from sensed inputs;

the processor configured to model a state estimation of the host vehicle and the target vehicle based on a set of speed and torque calculations about each vehicle, and by utilizing a first time vector of ACC torque and a second time vector of detection of a brake light of the target vehicle;

the processor configured to generate a set of results from at least one reward function based on one or more state estimations of the host and the target vehicle;

the processor configured to process the set of results with driver behavior data established by RL to correlate one or more control actions to the driver behavior data; and

wherein the process is further configured to generate the set of results from a plurality of reward functions comprising:

a first reward function R 1 pertaining to driver feedback and based on a difference between a driver torque (τ Driver ) and an ACC torque (τ ACC );

a second reward function R 2 pertaining to an indication and based on the first time vector of ACC torque ( τ Acc ) and the second time vector of detection of the brake light of the target vehicle ( b );

a third reward function R 3 pertaining to distance and based on a distance (d) between the host vehicle and the target vehicle; and

a fourth reward function R 4 pertaining ACC performance and based on an ACC speed ((v ACC ) and an estimated speed (v est ).

11. The system of claim 10 , further comprising the processor configured to:

apply at least one control action correlated to the driver behavior data established by RL to adjust at least one control action of the ACC system of the host vehicle;

adjust the at least one control action that is correlated to the driver behavior data established by RL based on a control safety check;

adjust the at least one control action that is correlated to the driver behavior data of the NN based on a control safety check; and,

calculate the reward function using a set of parameters comprising speed and torque calculations of estimations in acceleration and speed of the host and target vehicles.

12. The system of claim 11 , further comprising:

the processor configured to adjust one or more distances between the host vehicle and the target vehicle based on learned driver behavior contained in the data of a learning matrix based on the generated set of results from the at least one reward function to create a profile of driver behavior.

13. The system of claim 12 , wherein the control safety check comprises a difference in speed between a safe speed and an estimation of the speed of the target and host vehicles.

14. A vehicle apparatus comprising:

a vehicle controller comprising a processor wherein the processor is coupled to a dynamic neural network (DNN) adapted with a Reinforcement Learning (RL) based on driver's feedback and configured to:

execute, by the processor, adaptive cruise control to receive a set of vehicle inputs about a host vehicle's operating environment and current operations;

identify, by the processor, a target vehicle operating in the host vehicle environment and quantifying a set of target vehicle parameters about the target vehicle derived from sensed inputs;

model, by the processor, a state estimation of the host vehicle and the target vehicle by generating a set of speed and torque calculations about each vehicle, and by utilizing a first time vector of ACC torque and a second time vector of detection of a brake light of the target vehicle;

generate, by the processor, a set of results from at least one reward function based on one or more modeled state estimations of the host and the target vehicle; and

correlate, by the processor, the set of results with driver behavior data adapted by RL to one or more control actions to the driver behavior data;

wherein the generating of the set of results is performed from a plurality of reward functions comprising:

a first reward function R 1 pertaining to driver feedback and based on a difference between a driver torque (τ Driver ) and an ACC torque (τ ACC );

a second reward function R 2 pertaining to an indication and based on the first time vector of ACC torque ( τ Acc ) and the second time vector of detection of the brake light of the target vehicle ( b );

a third reward function R 3 pertaining to distance and based on a distance (d) between the host vehicle and the target vehicle; and

a fourth reward function R 4 pertaining ACC performance and based on an ACC speed ((v ACC ) and an estimated speed (v est ).

15. The vehicle apparatus of claim 14 , further comprising:

the processor configured to:

apply at least one control action that is correlated to the driver behavior data adapted by RL to adjust at least one operation of the adaptive cruise control of the host vehicle.

16. The vehicle apparatus of claim 15 , further comprising:

the processor configured to:

adjust the at least one control action that is correlated to the driver behavior data adapted by RL based on a control safety check.

17. The vehicle apparatus of claim 16 , further comprising:

the processor configured to:

update data of a learning matrix based on the generated set of results from the at least one reward function to create a profile of driver behavior.

18. The vehicle apparatus of claim 17 , further comprising:

the processor configured to:

calculate the reward function using a set of parameters comprising speed and torque calculations of estimations in acceleration and speed of the host and target vehicles.

19. The vehicle apparatus of claim 18 , further comprising:

the processor configured to:

adjust one or more distances between the host vehicle and target vehicle based on learned driver behavior contained in the data of the learning matrix.

20. The vehicle apparatus of claim 14 , wherein:

the first reward function R1 is calculated in accordance with the following equation R 1 =(τ ACC −τ Driver ) −2 ;

the second reward function R2 is calculated in accordance with the following equation R 2 =Convolution( τ Acc * b );

the third reward function R3 is calculated in accordance with the following equation R 3 =d 2 ; and

the fourth reward function R4 is calculated in accordance with the following equation R 4 =(v ACC −v est ) −2 .