Vehicle control system having an adaptive controller

• US 7,177,743 B2
• Filed: 06/02/2004
• Issued: 02/13/2007
• Est. Priority Date: 06/02/2003
• Status: Expired due to Fees

• Current Assignee: Toyota Motor Corporation
• Original Assignee: Toyota Motor Engineering & Manufacturing North America Incorporated (Toyota Motor Corporation)
• Inventors: Roy, Rini
• Primary Examiner(s): NGUYEN, CUONG H
• Application Number:
  US10/859,040
• Publication Number:
  US 20040243292A1
• Time in Patent Office: 986 Days
• Field of Search: 701/36, 701/40, 701/42, 701/44, 701/58, 701/60, 706 15- 16, 706/23, 706 26- 27
• US Class Current: 701/36
• CPC Class Codes: B60W 2050/0057 Frequency analysis, spectral techniques or transforms
• CPC Class Codes: B60W 50/06 Improving the dynamic response of the control system, e.g. improving the speed of regulation or avoiding hunting or overshoot
• CPC Class Codes: G01S 2013/932 using own vehicle data, e.g. ground speed, steering wheel direction
• CPC Class Codes: G01S 2013/9322 using additional data, e.g. driver condition, road state or weather data

Abstract

The vehicle control system having an adaptive controller is provided that accomplishes unsupervised learning such that no prior extensive training is needed for every situation. The inventive controller system is based on a neural network evolved with genetic algorithm. The genetic algorithm will determine the parameters of the neurons, the connections between the neurons and the associated weights to yield the best results. The genetic algorithm evaluates current candidate structures for accomplishing the desired result and develops new candidate structures by reproducing prior candidate structures with modification that replaces the least fit former candidate structures until the system is well satisfied. The vehicle control system is well satisfied when the desired result is met or some failure condition is triggered for vehicle control system whereby the action is never repeated.

Claims

1. An artificial intelligence control system for use in dynamically controlling at least one vehicle system comprising:

a plurality of sensors disposed on the vehicle operative to detect and input driving data into the system;

at least one network including a plurality of data processing nodes having weighted interconnections wherein each node is predisposed with a vehicle performance parameter and operative to receive inputs from a portion of said plurality of sensors and other interconnected nodes, and wherein said interconnections are weighted according to importance for establishing a first candidate structure that accomplishes a desired result relative to said vehicle performance parameter;

a feedback channel operative to communicate result data to at least one network whereby said result data is compared to said vehicle performance parameter to determine whether the result data satisfies the desired result; and

a genetic algorithm employed by said at least one network and operative to create a population of candidate structures derived from said first candidate structure, said genetic algorithm further operative to iteratively perform predetermined calculations on said population of candidate structures in order to determine at least one other candidate structure operative to produce said desired result, and wherein said genetic algorithm is operative to store said at least one other candidate structure when said desired result is accomplished.

2. The system of claim 1 wherein the genetic algorithm is adapted to coordinate operations of a plurality of interconnected networks working together to accomplish a common task.

3. The system of claim 1 wherein the genetic algorithm is further operative to adjust said weighted interconnections whereby said network can implement said at least one other candidate structure.

4. The system of claim 1 wherein the genetic algorithm controls the system to discard said at least one other candidate structure that accomplishes an undesired result.

5. The system of claim 1 wherein the genetic algorithm is operative to control the system to replace least fit candidate structures with candidate structures that better satisfy said desired result.

6. The system of claim 1 wherein the genetic algorithm is operative to perform an evaluation function for determining how well said at least one other candidate structure satisfies said desired result.

7. The system of claim 1 wherein the genetic algorithm is operative to determine said vehicle performance parameter within said nodes and said weighted interconnections between said nodes whereby a candidate solution that satisfies said desired result is established.

8. The system of claim 1 wherein each of said nodes further comprises at least one output for transmitting a control signal to the at least one vehicle system in response to receiving said inputs.

10. The system of claim 1 further comprising a coordinating controller operative to produce a signal that causes said genetic algorithm to iteratively perform said predetermined calculations.

11. The system of claim 10 wherein said coordinating controller produces said signal at 25 ms intervals.

9. An artificial intelligence control system for use in dynamically controlling at least one vehicle system comprising:

a plurality of sensors disposed on the vehicle operative to detect and input driving data into the system;

at least one network including a plurality of data processing nodes having weighted interconnections wherein each node is predisposed with a vehicle performance parameter and operative to receive inputs from said plurality of sensors and other interconnected nodes, and wherein said interconnections are weighted according to importance for establishing a first candidate structure that accomplishes a desired result relative to said vehicle performance parameter, said nodes further including at least one output for transmitting a control signal to the at least one vehicle system in response to receiving said inputs;

a feedback channel operative to communicate result data to said at least one network whereby said result data is compared to said vehicle performance parameter to determine whether the result data satisfies said desired result; and

a genetic algorithm employed by said at least one network and operative to create a population of candidate structures derived from said first candidate structure, said genetic algorithm further operative to iteratively perform predetermined calculations on said population of candidate structures in order to determine at least one other candidate structure operative to produce said desired result, and wherein said genetic algorithm is operative to store said at least one other candidate structure when said desired result is accomplished.