Method for determining the driving limits of a vehicle

• US 8,599,005 B2
• Filed: 06/20/2007
• Issued: 12/03/2013
• Est. Priority Date: 06/23/2006
• Status: Expired due to Fees

• Current Assignee: Nodbox Co.
• Original Assignee: Nodbox Co.
• Inventors: Fargas, Thierry, Clarac, Dominique
• Primary Examiner(s): Bugg, George
• Assistant Examiner(s): OBINIYI, PAULSON IDOWU
• Application Number:
  US12/306,442
• Publication Number:
  US 20090201140A1
• Time in Patent Office: 2,358 Days
• Field of Search: 340/441, 340/438, 340/425.5, 340/440, 180/282, 180/179, 180/197, 701/38, 701/1, 701/93, 701/301
• US Class Current: 340/441
• CPC Class Codes: B60W 2050/0057 Frequency analysis, spectral techniques or transforms
• CPC Class Codes: B60W 2050/0075 Automatic parameter input, automatic initialising or calibrating means
• CPC Class Codes: B60W 2556/50 of positioning data, e.g. GPS [Global Positioning System] data
• CPC Class Codes: B60W 40/101 Side slip angle of tyre
• CPC Class Codes: B60W 40/103 Side slip angle of vehicle body

Abstract

A method for making it possible to set the link between all the types of vehicles and roads with the rolling limits on the roadway is described. This setting can be established from the existing road data bases and from the characteristics of the known vehicles. This method is capable of determining the vehicle rolling limits. A device which can be fitted on any vehicle and capable of implementing the method according to the invention is also disclosed.

Claims

1. A method for determining at least one rolling limit of a vehicle moving on a given traffic lane of a roadway, comprising:

acquiring data characterising plural vehicles, where said data are organised by families each representative of vehicles sharing at least one vehicle characterising parameter, and where a given vehicle belongs to one of the families,

acquiring geometric data defining a sample of plural traffic lanes representative of traffic lanes comprising the given traffic lane,

acquiring data characterising the sample of plural traffic lanes,

learning a neural network with the acquired data characterising plural vehicles, with the acquired geometric data and with the acquired data characterising the sample,

obtaining geometric data defining the given traffic lane,

obtaining data characterising the given traffic lane,

obtaining data defining a vehicle path,

obtaining data characterising the given vehicle,

determining, in real time, a value for the rolling limits of the given vehicle, comprising;

providing the obtained geometric data, the obtained data charactering the given traffic lane, the obtained data defining the vehicle path and the obtained data characterising the given vehicle in input of the neural network,

calculating the rolling limit with the neural network based on the obtained geometric data, the obtained data charactering the given traffic lane, the obtained data defining the vehicle path and the obtained data characterising the given vehicle, and

providing the calculated rolling limit in output of the neural network.

2. The method according to claim 1, wherein the rolling limits are selected among a rollover speed, a rollover acceleration, a slide-slip speed and a slide-slip acceleration.

3. The method according to claim 1, wherein the learning by the neural network includes:

a step for setting between vehicles characterising data, geometric and characterising simplified data of the traffic lane and the rolling limits.

4. The method according to claim 1, wherein the learning by the neural network includes:

a step of setting between the vehicle characterising data, permissible path data for the vehicle with respect to the traffic lane and the rolling limits.

5. The method according to claim 1, wherein the learning by the neural network includes:

a step of setting between vehicle characterising, data characterising vehicles in motion, a speed of a centre of gravity with respect to X and Y, a rate of yaw, an acceleration of the centre of gravity with respect to X and Y, a drift angle, a steering of a wheel, a speed of the steering of the wheel and the rolling limits.

6. The method according to claim 1, wherein

particular points are defined on the given traffic lane and a position thereof is defined,

complex relations between the data characterising the given traffic lane, the geometric data defining the given traffic lane, a permissible dynamic path and the particular points are defined,

the motion of the given vehicle is measured, an existence of particular points met during the motion is deduced and a relative path of said vehicle is determined,

measured data of the motion are compared to the position of the particular points met upon a passage of the vehicle at said particular points so as to increase an accuracy of a vehicle location.

7. The method according to claim 6, wherein the particular points are defined by curvature or angle or altitude changes criteria which are remarkable.

8. The method according to claim 1, wherein the geometry data defining the traffic lane, the data characterising the traffic lane, the data defining the vehicle path, the data characterising the vehicle and variables of the rolling limits are obtained by:

simulation,

or real measures,

or a vehicle manufacturer,

or an independent organisation communicating qualification values,

or through transmission of information delivered by map data suppliers,

or by processing video images or real time or delayed time observation from satellites.

9. The method according to claim 1, wherein:

the rolling limit values are compared with instantaneous values of the rolling of the vehicle;

a warning is generated for a driver when an instantaneous value is greater than a matching limit value.

10. The method according to claim 1, wherein at least one indicator accessible by a vehicle driver is generated to inform him/her:

of a danger in the paths,

or an improved position,

or of an existence of a relief or slope.

11. The method according to claim 1, wherein at least one indicator about an incidence on safety of an instantaneous speed of the vehicle at a given spot and downstream the traffic lane is generated in real time to a driver, relative to at least one of the following parameters:

rolling limit, his or her path, a regulatory speed, the data characterising the traffic lane and weather conditions.

12. The method according to claim 1, wherein the data characterising the sample of plural traffic lanes include at least one of the following criteria:

an adherence parameter;

data indicating danger;

data indicating of an obligation of a low velocity passage.

13. The method according to claim 12, wherein the rolling limits include at least one of the following limits:

braking limit, acceleration limit, travelling limit;

instruction on the opposite lane, ram-off roadway accident, frontal collision with a vehicle, rear collision with a vehicle, collision with an engineering structure, a collision with an obstacle on the road, collision with pedestrians;

global adherence, rolling characteristic of the road relative to onboard equipment, road characteristics relative to a passengers'"'"' comfort, road characteristics relative to a driver'"'"'s fatigue;

defects and failures;

driving, driver, vehicle and vehicle equipment, selection of adjustment modes, variables to be corrected.

14. The method according to claim 2, wherein the rolling limits include traffic limits selected among:

a traffic slowdown, driving up a traffic line, a traffic flow.

15. The method according to claim 14, wherein the learning of the neural network includes:

a step of setting between geometric data, map data, crossing road, types of vehicles driving on the road, signalling, variable signs, alternate street lights, number of types of vehicles driving on these roads, the speed of the vehicle types driving on these roads, simultaneous weather conditions, simultaneous ephemerides, simultaneous events on and off the road, simultaneous accidentology, and the rolling limits.

16. The device for determining the rolling limits of a vehicle moving on the traffic lane of the roadway implementing the method according to claim 1.

17. The device according to claim 16 including a circuit of an in situ programmable gate network type for a strongly parallel processing of data.

18. The method according to claim 2, wherein the learning by the neural network includes:

a step for setting between vehicles characterising data, geometric and characterising simplified data of the traffic lane and the rolling limits.

19. The method according to claim 2, wherein the learning by the neural network includes:

a step of setting between the vehicle characterising data, permissible path data for the vehicle with respect to the traffic lane and the rolling limits.

20. The method of claim 1, wherein the at least one vehicle characterising parameter comprises at least one of a vehicle drive type and a weight/power output ratio.

21. The method of claim 1, wherein data characterising the given vehicle comprises at least one of a vehicle drive type, a weight/power output ratio, a dimension, a mass and a wheel base.

22. A method for determining at least one rolling limit of a given vehicle moving on a given traffic lane of a roadway, comprising:

acquiring data characterising plural vehicles, where said data are organised by families each representative of vehicles sharing at least one vehicle characterising parameter, and where the given vehicle belongs to one of the families,

acquiring geometric data defining a sample of plural traffic lanes representative of traffic lanes comprising the given traffic lane,

acquiring data characterising the sample of plural traffic lanes,

learning a neural network with the acquired data characterising plural vehicles, with the acquired geometric data and with the acquired data characterising the sample,

obtaining geometric data defining the given traffic lane,

obtaining data characterising the given traffic lane,

obtaining data defining a vehicle path,

obtaining data characterising the given vehicle,

determining, in real time, a value for the rolling limit of the given vehicle comprising;

providing the obtained geometric data, the obtained data charactering the given traffic lane, the obtained data defining the vehicle path and the obtained data characterising the given vehicle in input of the neural network,

calculating the rolling limit with the neural network based on the obtained geometric data, the obtained data charactering the given traffic lane, the obtained data defining the vehicle path and the obtained data characterising the given vehicle, and

providing the calculated rolling limit in output of the neural network, and

refining the data defining the vehicle path comprising;

i) defining particular points on the given traffic lane and defining a position thereof,

ii) defining relations between the data characterising the given traffic lane, the geometric data defining the given traffic lane, permissible dynamic paths and the particular points,

iii) measuring a motion of the given vehicle,

iv) deducing an existence of particular points met during the motion,

v) determining a relative path of said vehicle among the permissible dynamic paths based on the particular points met during the motion.