ROBOT FOR DRIVING A MOTOR VEHICLE, COMPRISING A CONTROL LIMITING SPEED VARIATIONS, VEHICLE AND METHOD AND PROGRAM BASED ON SUCH A ROBOT

Abstract

The invention relates to a robot for driving a motor vehicle, comprising:- a driving regulator producing a speed setpoint (Csg);- at least one driving computer comprising:- a means for analyzing the speed of the vehicle (VV) in driving cycles;- a means for detecting (D) stable zones where the speed setpoint (Csg) is constant for a given time;- a means for storing commands from the driving regulator for each stable zone;- a means for applying average or median commands from the driving regulator corresponding to the new stable zone. The invention also relates to a vehicle, a method and a program based on such a robot. Figure 2

Description

Title of the invention: ROBOT FOR DRIVING A MOTOR VEHICLE, COMPRISING A CONTROL LIMITING SPEED VARIATIONS, VEHICLE AND METHOD AND PROGRAM BASED ON SUCH A ROBOT

[0001] The invention relates to the field of motor vehicle driving robots, in particular those used on roller bench (RB) type test systems.

[0002] The invention relates more particularly to the configuration of the instructions of these robots.

[0003] There are driving robots composed of actuators capable of mechanically controlling the brake pedal, the accelerator pedal and the gearshift lever, with a specific operating method.

[0004] The invention is based more particularly on the speed instruction monitoring carried out by the applicant's robot described in patent FR3100190B1.

[0005] The driving cycle tracking on a stabilized profile carried out by the applicant's driving robot adversely impacts the driving cycle validation index due to small speed oscillations observed in the cycle.

[0006] An objective of the present invention is to remedy the defects of the prior art, and in particular to propose a solution for limiting speed oscillations when driving by a driving robot, and improving the cycle variation index.

[0007] To achieve this objective, the invention proposes a robot for driving a motor vehicle, comprising: - a driving regulator implementing at least one speed instruction, the driving regulator comprising a global regulator supervising an accelerator regulator and a brake regulator; - at least one driving computer; said computer comprising: - an analysis means analyzing the speed of the vehicle during driving cycles of a test; - a means of detecting stable zones detecting time zones of a driving cycle where the speed setpoint is constant for a given time; - a command storage means recording the command values ​​of the driving regulator for each stable zone; - a means of applying control implementing, in the event of a new stable zone, an average or median control value of the driving regulator corresponding to the new stable zone.

[0008] Advantageously, the invention makes it possible to optimize the regulation of the setpoint tracking carried out by the robot in the case of stabilized setpoint profiles. Optimizing the setpoint tracking makes it possible to eliminate speed oscillations, even small ones, to improve the rating index used to validate the test results.

[0009] Preferably, the given time during which the setpoint is constant is between 5s and 15s, and is preferably 10s.

[0010] This makes it possible to quickly determine whether the speed setpoint is constant.

[0011] Preferably, the speed setpoint is constant if its value at a previous instant and at a following instant are equal.

[0012] This makes it possible to efficiently determine whether the speed setpoint is constant.

[0013] Preferably, the interval between the previous instant and the following instant is between 10 and 25s, and is preferably 15s.

[0014] This makes it possible to quickly determine whether the speed setpoint is constant.

[0015] Preferably, the speed setpoint is constant if a difference in values ​​between the speed setpoint and the speed is less than a speed threshold.

[0016] This makes it possible to efficiently determine whether the speed setpoint is constant.

[0017] Preferably, the speed threshold is between 0.5 km / h and 2 km / h, and is preferably 1km / h.

[0018] This makes it possible to determine precisely whether the speed setpoint is constant.

[0019] Preferably, the control application means, in the event of a new stable zone, corrects the control by the average or median control value of the driving regulator corresponding to the new stable zone, as well as a negative prepositioning component and / or a derived component of the regulator.

[0020] This makes it possible to limit speed oscillations and reproduce human piloting.

[0021] The invention further relates to a motor vehicle comprising a driving robot according to the invention.

[0022] Another subject of the invention relates to a method for automatically driving a motor vehicle by means of a driving robot according to the invention, characterized in that it comprises the following steps: - an analysis step in which the vehicle speed is analyzed during driving cycles of a test; - a stable zone detection step in which time zones of a driving cycle are detected where the speed setpoint is constant for a given time; - a command storage step in which the command values ​​of the driving regulator are recorded for each stable zone; - a control application step in which, in the case of a new stable zone, an average or median control value of the regulator is implemented. conduct corresponding to the new stable zone.

[0023] The invention further relates to a computer program comprising program code instructions for executing the steps of the driving method according to the invention, when said program operates on a computer.

[0024] The invention will be further detailed by the description of non-limiting embodiments, and on the basis of the appended figures illustrating variants of the invention, in which: - [Fig.l] illustrates an overall diagram of the driving robot connected to the brake and accelerator of a motor vehicle; - [Fig.2] schematically illustrates changes in commands, instructions and detection of a new stabilized zone; - [Fig.3] illustrates developments following those of [Fig.2] and a detection of the end of the new stabilized zone; - [Fig.4] schematically illustrates evolutions similar to those of figures 2 and 3, with variations in speed in the prior art and the absence of such variations in the invention; - [Fig.5] schematically illustrates an automatic driving method according to a preferred embodiment of the invention.

[0025] The invention proposes in particular to integrate into the applicant's driving robot. The invention proposes to drive a vehicle equipped with a so-called development computer and an automatic transmission on a roller bench. This makes it possible to follow in complete autonomy a speed setpoint generally resulting from validation or homologation cycles by carrying out a detection of stabilized zones (constant setpoint) and by applying in this zone a specific regulation mode making it possible to reproduce human driving.

[0026] The problems solved by the invention are as follows: - Regulate the vehicle speed on a roller bench type test bench using an automation system whose installation and implementation time remain compatible with standard development processes regardless of the vehicle tested; - Obtain a quality of instruction monitoring compatible with test requirements, whatever the instruction template and the vehicle to be tested and reproduce human driving behavior; - Avoid speed oscillations on stabilized instructions (constant speed instruction) which impacts the driving index; - Avoid having to manage the configuration of the robot parameters according to different speed setpoint templates and the type of vehicle present for the test.

[0027] Without this invention, in the presence of a stabilized zone, the setpoint tracking undergoes a oscillation which impacts the validation index of the test.

[0028] The invention described here makes it possible to detect the zones where the setpoint remains constant in order to trigger a specific regulation mechanism capable of limiting speed oscillations and limiting the increase in the index.

[0029] The invention comprises two methods which complement each other:

[0030] The first method consists of analyzing the speed setpoint during the test. This analysis makes it possible to detect the time zones where the speed setpoint remains constant for a minimum amount of time in the past and in the future. This part of the cycle will be called the “stabilized” zone.

[0031] The second method uses the data produced in the first step. During the test, in the presence of a "stabilized" zone, this invention stores the control samples produced by the regulator for a configurable time and then calculates a new control from the median value of the saved samples and applies this value by deactivating the initial regulation of the robot. This method therefore imposes a fixed control, to obtain a stable speed centered on the setpoint. The improvement in driving monitoring induced by this method makes it possible to limit the increase in the validation index of the test.

[0032] [Fig.l] illustrates the driving robot according to the invention. The robot comprises a driving regulator and at least one driving computer C. In particular, the driving regulator comprises an accelerator regulator RA connected to the accelerator of the vehicle PA, a brake regulator RF connected to the brake of the vehicle PF, and a global regulator RG regulating the other two regulators RA, RF.

[0033] The invention is situated in [Fig.l] at the level of the global regulator RG. The processing is carried out during testing by reading the speed instructions Csg.

[0034] [Fig.2] describes the detection of the stabilized zone D(Stb) (we will speak of detection means D). The reference Cmd concerns the control values; VV concerns the vehicle speed values ​​in km / h, and Csg, the speed setpoint values. The reference I concerns the intensity values ​​of the control actuator.

[0035] In the first step, the stabilized zones are detected where the setpoint Csg remains stable for a minimum configurable time (called “given time”). The reading is carried out at each instant “t” on the previous setpoints Csgl(tX) and the future setpoints Csg2(t+Y). The detection of the stabilized zone is activated if: - Csgl = Csg2 ; and - the loop deviation “Eb” between “Speed ​​setpoint” and “speed” is less than a threshold that can be set in absolute value.

[0036] In [Fig.2], the stabilized zone detection state (1) is activated when the value of the setpoint Csgl(t-5s) is equal to the value of the setpoint Csg2(t+10s). In the preferred setting of the invention: X = 5; seconds Y = 10 seconds; and Eb = 1km / h.

[0037] It is important to keep a parameter X > 5 seconds to allow time for the RG regulator to converge towards a command adapted to the stabilized zone at the exit of the speed overflow.

[0038] When the detection of the stabilized zone is activated, the invention stores the control values ​​of the RG regulator for a few seconds to subsequently extract the median value. Increasing the number of samples taken into account for the calculation makes it possible to optimize the final control value. Currently this parameter is set to 10s. When this step is completed, the specific control state (2) is activated.

[0039] In the presence of state (2), the invention imposes the calculated command (namely the median value of the samples) instead of the standard calculation carried out by the RG regulator. [Fig.2] shows that the command flow (3) freezes at the moment of switching from state (2) to the calculated value.

[0040] [Fig.3] describes the end of the stabilized zone. When the stabilized zone ends, states (1) and (2) fall and the regulator control (3) is initialized to the standard pilot value.

[0041] To avoid too large a command jump, the integral component of the regulator is (re)calculated to have a command value equal to the median value at the time of returning to standard control mode, according to the formula: Cdez = CdeP - CdeFF + Cde^ + CdeMedianeavec Cde7: an integral component of the regulator; CdeP; the standard piloting value; CdeFF; a regulator prepositioning component; CdeD; a derivative component of the regulator; Cde^^,^, ; the median order calculated in the previous step.

[0042] For information, the standard command in the case of regulation by state feedback is calculated according to the formula: Cde = Cde7 - Cdep + CdeFF - Cde^with Order; the standard final order.

[0043] [Fig.4] shows a setpoint monitoring with and without the invention. As [Fig.4] indicates, the invention makes it possible to reduce oscillations on the control and consequently oscillations of ± 1 km / h on the speed obtained (absent in the figure below). The validation index of the autonomy cycles is consequently optimized.

[0044] The invention makes it possible to reproduce the human driving style on stabilized zones of instructions. Monitoring instructions with the invention makes it possible to improve the validation criterion.

Claims

Claims

1. Robot for driving a motor vehicle (V), comprising: - a driving regulator (RG, RA, RF) implementing at least one speed setpoint (Csg), the driving regulator comprising a global regulator (RG) supervising an accelerator regulator (RA) and a brake regulator (RF); - at least one driving computer (C); said computer (C) comprising: - an analysis means (AV) analyzing the speed of the vehicle (VV) during driving cycles of a test; - a stable zone detection means (DS) detecting time zones of a driving cycle where the speed setpoint (Csg) is constant for a given time; - a command storage means (SC) recording the command values ​​of the driving regulator (RG, RA, RF) for each stable zone;- a control application means (AC) implementing, in the event of a new stable zone, an average or median control value of the driving regulator (RG, RA, RF) corresponding to the new stable zone.;

2. Driving robot according to claim 1, characterized in that the given time during which the instruction is constant is between 5s and 15s, and is preferably 10s.

3. Driving robot according to any one of claims 1 to 2, characterized in that the speed setpoint (Csg) is constant if its value at a previous instant and at a following instant are equal.

4. Driving robot according to any one of claims 1 to 3, characterized in that the interval between the previous instant and the following instant is between 10 and 25s, and is preferably 15s.

5. Driving robot according to any one of claims 1 to 4, characterized in that the speed setpoint (Csg) is constant if a difference in values ​​between the speed setpoint (Csg) and the vehicle speed (VV) is less than a speed threshold.

6. Driving robot according to claim 5, characterized in that the speed threshold is between 0.5 km / h and 2 km / h, and is preferably 1 km / h.

7. A driving robot according to any one of claims 1 to 6, ca-

8.

9.

10. characterized in that the control application means (AC), in the event of a new stable zone, corrects the control by the average or median control value of the driving regulator (RG, RA, RF) corresponding to the new stable zone, as well as a negative prepositioning component and / or a derived component of the regulator. Motor vehicle comprising a driving robot according to any one of claims 1 to 7. Method for automatically driving a motor vehicle by means of a driving robot according to any one of claims 1 to 7, characterized in that it comprises the following steps: - an analysis step in which the vehicle speed (VV) is analyzed during driving cycles of a test; - a stable zone detection step in which time zones of a driving cycle are detected where the speed setpoint is constant for a given time; - a command storage step in which the driving regulator command values ​​(RG, RA, RF) are recorded for each stable zone; - a control application step in which, in the case of a new stable zone, an average or median control value of the driving regulator (RG, RA, RF) corresponding to the new stable zone is implemented. A computer program comprising program code instructions for executing the steps of the driving method according to claim 9, when said program operates on a computer.

Images