Method and system of road driving optimization with decoupling of vehicle status and traffic factors
A method and system of road driving optimization, having vehicle sensors configured to collect external sensor data, vehicle-state data, and communications data; a control module configured to analyze the collected data to detect a trigger event, a status of a target vehicle, an achievable speed range, and instant traction force; a decoupling estimator module configured to analyze the trigger event, the status of the target vehicle, the achievable speed range, and a personalized driver profile to determine a maximum free flow distance and an arrival speed at the free flow distance; a machine learning model configured predict a speed profile of the host vehicle approaching the trigger event based on the determined free flow distance, the determined arrival speed at the free flow distance, and the instant traction force; and a cruise control system configured to implement the predicted speed profile.
1 . A method of driving a host vehicle, comprising:
receiving, by a control module, data collected by at least one vehicle sensor;
analyzing the collected data, by the control module, to detect a plurality of trigger events and determine an achievable speed range for each of the trigger events;
determining, by a decoupling estimator module, a maximum free flow distance and an arrival speed at the maximum free flow distance based on the plurality of detected trigger events and the determined achievable speed range for each of the trigger events;
determining an instant traction force of the host vehicle;
inputting the determined maximum free flow distance, the determined arrival speed at the maximum free flow distance, and the instant traction force of the host vehicle into a pretrained machine learning (ML) model to predict a speed profile to arrive at the maximum free flow distance, wherein the predicted speed profile is configured to minimize a fuel consumption and a jerk motion of the host vehicle;
driving the host vehicle toward the maximum free flow distance based on the predicted speed profile;
updating a personalized operator profile with the predicted speed profile; and
continual training of the ML model in real-time based on the maximum free flow distance, the arrival speed at maximum free-flow distance, the instant traction force, and the updated personalized operator profile.
2 . The method of claim 1 , wherein the plurality of trigger events include a target vehicle, and the method further includes:
analyzing the collected data to determine a status of the target vehicle, wherein the determined status of the target vehicle includes a location of the target vehicle relative to the host vehicle and a speed of the target vehicle; and
wherein determining, by the decoupling estimator module, the maximum free flow distance and the arrival speed at the maximum free flow distance is further based on the determined status of the target vehicle.
3 . The method of claim 2 , further comprising reiterating the predicted speed profile by inputting incremental corrections of the determined maximum free flow distance and the determined arrival speed at the maximum free flow distance into the pretrained ML model.
4 . The method of claim 1 , wherein driving the host vehicle toward the maximum free flow distance based on the predicted speed profile is executed by an Adaptive Cruise Control (ACC) system.
5 . The method of claim 1 , wherein the determined free flow distance is greater than a predetermined free flow distance threshold.
6 . The method of claim 1 , wherein the pretrained machine learning (ML) model is trained on a deep neural network.
7 . The method of claim 1 , wherein the plurality of trigger events is include a dynamic traffic light, and wherein the collected data includes a current state of the dynamic traffic light and a time interval until a change to a next state of the dynamic traffic light.
8 . A tangible, non-transitory, machine-readable medium, comprising machine-readable instructions, that when executed by a processor, cause the processor to:
receive personalized operator profile, external sensor data, and vehicle-state data;
analyze the external sensor data and the vehicle-state data to determine an instant traction force, a plurality of trigger events, and an achievable speed range for each of the plurality of trigger events;
input the plurality of determined trigger events and the determined achievable speed range for each of the plurality of trigger events into a decoupling estimator to determine a maximum free flow distance and to determine an arrival speed at the maximum free flow distance;
input the personalized operator profile, the instant traction force, the determined maximum free flow distance, and the determined arrival speed at the maximum free flow distance into a machine learning model to predict a speed profile to arrive at the maximum free flow distance, wherein the predicted speed profile is configured to minimize a fuel consumption and a jerk motion of a host vehicle; and
drive the host vehicle toward the maximum free flow distance in accordance with the predicted speed profile.
9 . The tangible, non-transitory, machine-readable medium of claim 8 ,
wherein the plurality of trigger events includes a dynamic traffic light; and
wherein the communications data includes a current state of the dynamic traffic light and a time interval until a change to a next state of the dynamic traffic light.
10 . The tangible, non-transitory, machine-readable medium of claim 8 , wherein the tangible, non-transitory, machine-readable medium, further comprising machine-readable instructions, that when executed by the processor, causes the processor to:
input the predicted speed profile into the ML model to continual training of the ML model.
11 . The tangible, non-transitory, machine-readable medium of claim 8 , wherein the tangible, non-transitory, machine-readable medium, further comprising machine-readable instructions, that when executed by the processor, causes the processor to:
update the personalized driver profile with the predicted speed profile.
12 . A system for driving a host vehicle, comprising:
a data-base configured to store a retriable and updatable personalized driver profile;
a plurality of vehicle sensors configured to collect external sensor data, vehicle-state data, and communications data;
a control module configured to analyze the external sensor data, vehicle-state data, and communications data to detect a plurality of trigger events, an achievable speed range for each of the plurality of trigger events, and an instant traction force;
a decoupling estimator module configured to analyze the plurality of trigger events and the achievable speed range for each of the plurality of trigger events to determine a maximum free flow distance and an arrival speed at the maximum free flow distance;
a pretrained machine learning (ML) model configured predict a speed profile of the host vehicle approaching the maximum free flow distance based on the instant traction force, the maximum free flow distance, the determined arrival speed at the maximum free flow distance, and an updated personalized driver profile; and
a cruise control system configured to implement the predicted speed profile; and
wherein the predicted speed profile is configured to minimize a fuel consumption and a jerk motion of the host vehicle.
13 . The system of claim 12 , wherein the ML model includes one of a deep neural network, a convolutional deep neural network, a deep belief network, and a recurrent neural network.
14 . The system of claim 12 ,
wherein the plurality of trigger events includes a dynamic traffic light; and
wherein the communications data includes a current state of the dynamic traffic light and a time interval until a change to a next state of the dynamic traffic light.
15 . The system of claim 14 , wherein the decoupling estimator module comprises a statistical model configured to perform statistical analysis based on the plurality of trigger events, and the achievable speed range for each of the plurality of trigger events to determine the maximum free flow distance and the arrival speed at the maximum free flow distance within a predetermined confidence value threshold.