ADAPTIVE PASSENGER COMFORT ENHANCEMENT IN AUTONOMOUS VEHICLES

• US 20180022361A1
• Filed: 07/19/2016
• Published: 01/25/2018
• Est. Priority Date: 07/19/2016
• Status: Active Grant

• Current Assignee: Futurewei Technologies Incorporated (Huawei Investment & Holding Co., Ltd.)
• Original Assignee: Futurewei Technologies Incorporated (Huawei Investment & Holding Co., Ltd.)
• Inventors: Rao, Lei, Li, Jian, Hu, Ron-Chung
• Granted Patent: US 10,029,698 B2
• Time in Patent Office: Days
• CPC Class Codes: B60R 16/037 for occupant comfort , e.g. for automatic adjustment of appliances according to personal settings, e.g. seats, mirrors, steering wheel
• CPC Class Codes: B60R 16/0373 Voice control in general G10L
• CPC Class Codes: B60W 10/30 including control of auxiliary equipment, e.g. air-conditioning compressors or oil pumps
• CPC Class Codes: B60W 2050/0014 Adaptive controllers
• CPC Class Codes: B60W 2050/0075 Automatic parameter input, automatic initialising or calibrating means
• CPC Class Codes: B60W 2420/403 Image sensing, e.g. optical camera
• CPC Class Codes: B60W 2420/54 Audio sensitive means, e.g. ultrasound
• CPC Class Codes: B60W 2540/21 Voice
• CPC Class Codes: B60W 2540/215 Selection or confirmation of options
• CPC Class Codes: B60W 2720/10 Longitudinal speed
• CPC Class Codes: B60W 30/143 Speed control B60W30/16 takes precedence
• CPC Class Codes: B60W 30/18 Propelling the vehicle
• CPC Class Codes: B60W 50/0098 Details of control systems ensuring comfort, safety or stability not otherwise provided for
• CPC Class Codes: B60W 50/085 Changing the parameters of the control units, e.g. changing limit values, working points by control input
• CPC Class Codes: B60Y 2300/18 Propelling the vehicle
• CPC Class Codes: G05B 13/024 in which a parameter or coefficient is automatically adjusted to optimise the performance
• CPC Class Codes: G05B 13/028 using expert systems only

Abstract

A system and method for performing self-learning for adaptively achieving passenger comfort enhancement in an autonomous vehicle. The system comprises a plurality of sensor inputs. Each sensor input provides data representative of voice responses and image responses from a passenger in the vehicle. A controller is coupled to the plurality of sensor inputs. The controller generates and updates a reward function that includes a plurality of driving state transitions. The reward function is updated based on destination information and the data representative of the voice and image responses. The controller further generates a goal function that determines an optimized driving state transition updates the goal function based on the updated reward function and a previous goal function. The controller also generates a vehicle speed control signal, based on the updated goal function, to control the speed of the autonomous vehicle.

Claims

1. A system to perform self-learning for adaptively achieving passenger comfort enhancement in an autonomous vehicle, the system comprising:

a plurality of sensor inputs configured to provide data representative of voice and image responses of a passenger in the vehicle; and

a controller coupled to the plurality of sensor inputs, the controller configured to generate and update a reward function comprising a plurality of driving state transitions, the reward function updated based on a received destination and the data representative of the voice and image responses, the controller further configured to generate a goal function that determines an optimized driving state transition and to update the goal function based on the updated reward function and a previous goal function, the controller further configured to generate a vehicle speed control signal based on the updated goal function to control a vehicle speed.

2. The system of claim 1, wherein the optimized driving state transition comprises a maximized vehicle speed and the controller is further configured to assign weights to the maximized vehicle speed and to the updated goal function, the weights based on the data representative of the voice or image responses.

3. The system of claim 2, wherein each of the plurality of driving state transitions comprises a range of vehicle speeds and a range of road segment curvatures.

4. The system of claim 3, wherein the reward function comprises a reward function matrix of transition and non-transition indications indicating possibilities for transitioning from a first state to a second state.

5. The system of claim 4, wherein the controller is further configured to update each transition indication by a first reward function value based on the image response and by a second reward function value based on the data representative of the voice response.

6. The system of claim 5, wherein the first reward function value is less than the second reward function value with successive positive feedback collected from the data representative of the voice response.

7. The system of claim 1, wherein the controller is further configured to determine a plurality of road segments between a source and a destination, wherein the controller generates the vehicle speed control signal for each of the plurality of road segments based on the updated goal function for each road segment.

8. The system of claim 7, wherein the controller is further configured to update the vehicle speed control signal based on passenger comfort rating and passenger speed rating.

9. The system of claim 1, wherein the plurality of sensors comprise a microphone and an image sensor.

10. A method for adaptively updating passenger comfort enhancement in an autonomous vehicle, the method comprising:

determining a source-destination route based on a vehicle initial position, destination, and route information for the source-destination route, the route information comprising road curvature and road speed limits for each of a plurality of route segments of the source-destination route;

obtaining passenger feedback for each route segment during vehicle operation over the source-destination route, the passenger feedback comprising passenger audio data and passenger image data;

updating a reward function comprising a plurality of driving states, the updating based on a previous reward function and the passenger feedback;

updating a goal function for each route segment based on a previous goal function and an optimized driving state transition from a first driving state to a second driving state, wherein the optimized driving state transition is based on the updated reward function; and

dynamically updating a vehicle speed for each route segment based on the updated goal function.

11. The method of claim 10, wherein each driving state comprises a vehicle speed range and a road curvature range.

12. The method of claim 11, further comprising converting the passenger image data to a binary representation of a passenger facial expression, wherein a first binary representation of the passenger facial expression is assigned to a positive facial expression and a second binary representation of the passenger facial expression is assigned to a negative facial expression.

13. The method of claim 12, wherein obtaining the passenger audio data comprises an aural response from the passenger.

14. The method of claim 13, further comprising converting the aural response to a binary representation of the aural response, wherein a first binary representation of the aural response is assigned to a positive aural response and a second binary representation of the aural response is assigned to a negative aural response.

15. The method of claim 14, wherein a the driving state comprises a first range of vehicle speed and a first range of road curvature and the second driving state comprises a second range of vehicle speed and a second range of road curvature, the first driving state being different from the second driving state, the method further comprising:

setting the reward function to an initial value;

transitioning from the first driving state to the second driving state;

if the binary representation of the facial expression is the first binary representation in response to the transition, assigning a first weighting value to the reward function, the first weighting value greater than the initial value;

if the binary representation of the facial expression is the second binary representation in response to the transition, maintaining the reward function at a current value;

if the binary representation of the aural response is the first binary representation in response to the transition, increasing the reward function to a smaller of;

the current value plus a second weighting value or a predetermined integer multiplied by the first weighting value; and

if the binary representation of the aural response is the second binary representation in response to the transition, decreasing the reward function by a largest of;

the current value minus the second weighting value or a predetermined integer.

16. The method of claim 15, wherein updating the goal function comprises:

setting the goal function to an initial value;

assigning a source-destination route segment state to each of the plurality of route segments, wherein each route segment state comprises a speed limit for the route segment and a curvature range for the route segment; and

selecting the updated goal function based on a maximum goal function value determined from the plurality of driving states.

17. A non-transitory computer-readable storage medium that stores instructions for execution by processing circuitry of an autonomous vehicle to perform operations that:

generate a reward function comprising a plurality of driving state transitions;

update the reward function based on destination information and passenger feedback including data representative of voice and image responses;

generate a goal function that determines an optimized driving state transition;

update the goal function based on the updated reward function and a previous goal function; and

generate a vehicle speed control signal to control the autonomous vehicle speed based on the updated goal function.

18. The non-transitory computer-readable storage medium of claim 17, wherein the operations determine a plurality of route segments between a source and a destination, each route segment comprising an assigned state having a speed limit and a road curvature for each respective route segment.

19. The non-transitory computer-readable storage medium of claim 18, wherein the operations further update the vehicle speed control signal for each of the plurality of route segments.

20. The non-transitory computer-readable storage medium of claim 19, wherein the operations further updates the vehicle speed control signal based on passenger comfort rating and speed rating.