METHOD FOR CONTROLLING A VEHICLE MOUNTED ON A ROLLER TEST BENCH BY A DRIVING ROBOT AND DRIVING ROBOT

Abstract

The method is implemented in a driving robot (1) managing the tracking by a vehicle (2) mounted on a roller test bench of a rolling speed setpoint (CSi) provided by a template stored in memory (GA_V), the method being of the type in which an anticipation time offset is introduced into the rolling speed setpoint. According to the invention, the method comprises recurring steps, carried out during a test cycle, of: a) measuring in real time a rolling index called "IWR", and b) then adjusting a current value of the anticipation time offset as a function of an error between the rolling index measured in step a) and a predetermined target rolling index. Figure 1

Description

Title of the invention: METHOD FOR CONTROLLING A VEHICLE MOUNTED ON A ROLLER TEST BENCH BY A DRIVING ROBOT AND DRIVING ROBOT

[0001] The present invention relates generally to the robotic control of a vehicle mounted on test equipment. More particularly, the invention relates to a method of piloting a vehicle mounted on a roller test bench using a driving robot, and to a driving robot implementing the aforementioned method. The invention is applicable, but not exclusively, to the development and type-approval testing of electrified vehicles of the all-electric, hybrid, and fuel cell types.

[0002] Vehicle driving robots are commonly used for conducting test cycles on motor vehicles, particularly roller test cycles. Thus, in the prior art, driving robots comprising actuators capable of mechanically moving vehicle control components, such as the brake and accelerator pedals, are known. A vehicle driving robot, disclosed by the applicant in its French patent application FR3100190A1, is also known. In this driving robot, acceleration commands are executed through communication between a computer in the robot and a computer in the vehicle, thereby eliminating the need for an actuator coupled to the vehicle's accelerator pedal.

[0003] Roller bench test cycles are performed by automotive manufacturers, particularly to comply with regulatory homologation procedures such as the WLTP (Worldwide Harmonized Light Vehicles Test Procedure). In the WLTP procedure, the homologation route for the electric range of an electrified vehicle includes phases with significant deceleration and acceleration gradients. The inventive entity has observed difficulties with current driving robots in achieving optimal route following during these phases of significant deceleration / acceleration gradients. The driving indices taken into account for validating the WLTP test cycles are impacted, in particular the IWR (Inertial Work Ratio), the recorded values ​​of which may be close to, or even outside, the acceptable limits for the test cycle.

[0004] Document FR3107366A1 describes a method for piloting a motor vehicle driving robot designed for type-approval testing. The robot is coupled to the vehicle and includes a controller into which a template is loaded. The speed setpoint for a route the vehicle must follow is defined in this method. In this process, the speed setpoint template is applied with an anticipation, specifically, with a time lag AT that corresponds to an average delay observed when the cruise control system follows the setpoint. Thus, a setpoint intended to be applied at time t is actually applied earlier, at time t-AT, in order to improve the cruise control system's adherence to the speed setpoint template.

[0005] It is desirable to propose a method and a vehicle driving robot that are capable of guaranteeing optimal route following for electrified vehicles, with satisfactory driving indices, even in the presence of significant deceleration / acceleration gradients.

[0006] According to a first aspect, the invention relates to a control method implemented in a driving robot that manages the tracking by a vehicle mounted on a roller test bench of a rolling speed setpoint provided by a template stored in memory, the method being of the type in which a time lag of anticipation is introduced into the rolling speed setpoint. According to the invention, the method comprises recurring steps, carried out during a test cycle, of: a) measuring in real time a rolling index called "IWR", and b) then adjusting a current value of the time lag of anticipation as a function of an error between the rolling index measured in step a) and a predetermined target rolling index.

[0007] According to a particular feature of the process, in step b), the current value of the anticipation time lag is adjusted by subtracting or adding to it a fixed time parameterized according to whether the aforementioned error has a positive or negative sign.

[0008] According to another particular feature of the method, with the template stored in memory being sampled and quantized with a predetermined step, step a) is carried out during a first time interval and step b) is carried out during a subsequent time interval.

[0009] According to yet another particular feature of the process, the current value of the anticipation time lag is limited between a predetermined maximum duration and a predetermined minimum duration.

[0010] The invention also relates to a vehicle driving robot mounted on a roller test bench, the driving robot comprising a computer, an actuator control device, an actuator assembly coupled to the vehicle, and a digital interface with the vehicle. According to the invention, the computer includes a memory storing program instructions for implementing the method briefly described above when these program instructions are executed by a processor in the computer.

[0011] The invention also relates to an assembly comprising a vehicle mounted on a roller test bench and a driving robot installed in the vehicle, in which the driving robot is a driving robot as briefly described above.

[0012] Other advantages and features of the present invention will become more apparent upon reading the detailed description below of several particular embodiments of the invention, with reference to the accompanying drawings, in which:

[0013] Fig. 1 is a general block diagram schematically showing a functional architecture of a particular embodiment of a driving robot according to the invention coupled to a vehicle mounted on a roller test bench.

[0014] Fig. 2 shows a logic diagram of a processing operation implemented by the method of the invention in a robot for driving an electrified vehicle.

[0015] Fig. 3 shows, by way of example, curves of a rolling index "IWR" calculated in real time, of a rolling speed setpoint and of an acceleration / deceleration command recorded in the driving robot of Fig. 2.

[0016] Figure 4 shows illustrative curves of results obtained with the process of the invention.

[0017] Figure [Fig. 5] shows illustrative curves of results obtained with the previous technique.

[0018] With reference to [Fig.1], the general architecture and operation of a particular embodiment 1 of a driving robot according to the invention is described below.

[0019] In this embodiment, the driving robot 1 is installed in a vehicle 2. The vehicle 2 is mounted on a roller test bench (not shown), typically for one or more test cycles. The driving robot 1 controls the vehicle 2 so as to make it follow a speed setpoint template that is stored in memory and is specific to the test cycle.

[0020] As seen in [Fig.1], the driving robot 1 essentially comprises a computer 10, an actuator control device 11, an actuator assembly 12 and a digital interface 13.

[0021] The computer 10 houses a ROB driving robot software module, located in a MEM memory of the computer. The ROB driving robot software module essentially comprises a C_CS setpoint correction software sub-module and an R_GL speed control software sub-module.

[0022] The C_CS software sub-module is specific to the invention. It enables the implementation of the method according to the invention by the execution of program code instructions by a processor (not shown) of the computer 10.

[0023] The software submodule C_CS implements a speed setpoint correction device (also referred to hereafter as C_CS). The software submodule C_CS is responsible for correcting an initial speed setpoint CSi from a taxiing speed setpoint template GA_V of the current test cycle, by introducing an anticipation time offset. The software submodule C_CS receives as input the initial speed setpoint CSi and taxiing information IN_R representative of the current test cycle and outputs a corrected speed setpoint CSa, which is provided as input to the speed control software submodule R_GL.

[0024] As described in more detail below, according to the invention, the corrected speed setpoint CSa includes an anticipation time offset that is adjusted according to the roll index “IWR” defined by the SAE J2951 standard (for “Inertial Work Ratio”). In the invention, the roll index “IWR” is calculated in real time by exploiting the aforementioned roll information IN_R.

[0025] The R_GL speed control software submodule implements a speed control device (also referred to as R_GL hereafter). The R_GL speed control device is responsible for the overall monitoring of the corrected speed setpoint CSa. The R_GL speed control device performs a speed control loop and provides a control command CR. The control command CR comprises an acceleration setpoint AC and a deceleration setpoint DE, which respectively control the increase and decrease of the vehicle 2's travel speed. The AC and DE setpoints are calculated from an error between the corrected speed setpoint CSa and a measured travel speed VV of vehicle 2. The AC and DE setpoints are transmitted respectively to control regulators R_AC and R_DE of the actuator control device 11.The measured driving speed VV is provided to the speed control device R_GL, via the digital interface 13, by a supervisory computer (not shown) of the vehicle 2.

[0026] The control regulators R_AC and R_DE are responsible for the servo control of the actuator assembly 12. In this example, the actuator assembly 12 comprises an accelerator pedal actuator A_PA controlled by regulator R_AC and a brake pedal actuator A_PF controlled by regulator R_DE. Actuators A_PA and A_PF are typically electrically controlled cylinder actuators. Actuators A_PA and A_PF are mechanically coupled to the accelerator pedal and brake pedal, respectively.

[0027] The R_AC controller converts the received acceleration setpoint AC into an accelerator pedal position setpoint P_PA. The R_AC controller performs a control loop to drive the actuator A_PA with the P_PA setpoint. The R_AC controller produces an actuator command C_PA, typically a pulse-width modulation (PWM) signal. This is based on an error between the pedal position setpoint P_PA and a pedal position feedback information R_PA provided by the actuator A_PA. The controller R_AC provides an APA position acknowledgment information for the accelerator pedal to the speed control device R_GL.

[0028] The controller R_DE converts the received deceleration setpoint DE into a brake pedal position setpoint P_PF. The controller R_DE establishes a control loop to drive the actuator A_PF with the setpoint P_PF. The controller R_DE generates an actuator command C_PF, typically a pulse-width modulated signal, based on an error between the pedal position setpoint P_PF and a pedal position feedback signal R_PF provided by the actuator A_PF. The controller R_DE provides a position acknowledgment signal ADE for the brake pedal to the speed control device R_GL.

[0029] With particular reference to [Fig.2], the processing carried out by the C_CS software sub-module for the implementation of the method of the invention is now described below.

[0030] The taxiing speed command template GA_V is a sampled and digitized template, with a sampling and quantization step PQ which is considered here to have a duration of one hundred milliseconds, PQ = 100 ms. The template GA_V is therefore composed of a plurality of N successive time intervals IT, IT0, ..., IT(n_i), ITn, IT(n+i), ..., ITn, each having a duration equal to the step PQ = 100 ms. The N time intervals IT are indexed from 0 to N, with the index n corresponding to any IT interval between 0 and N. Thus, during any time interval ITn, the template GA_V provides a corresponding taxiing speed command CSi(n).

[0031] Generally, according to the invention, the anticipation time offset introduced in the rolling speed setpoint is adjusted according to the rolling index IWR which is calculated in real time by the processing method.

[0032] The SAE J2951 standard defines the rolling resistance index by the following equation:

[0033] IWR = 100*(WRD-WRT) / WRT, with WRT corresponding to the kinetic energy of the vehicle traveling at the set speed and WRD corresponding to the kinetic energy of the vehicle traveling at the actual speed obtained.

[0034] The rolling resistance index IWR thus represents, in %, a relative difference in kinetic energy to bring the vehicle back to the target speed (by an acceleration or a deceleration).

[0035] Curves C1, C2, and C3 in [Fig. 3] illustrate the evolution over time (t) of a rolling resistance index (IWR), a rolling speed setpoint (CS), and an acceleration / deceleration control (CAD), respectively. The rolling resistance index (IWR), calculated in real time, varies substantially between -1 and 2.5 percent (%). The rolling speed setpoint (CS) is indicated in km / h. The acceleration / deceleration control CAD is indicated as a percentage (%), with a plus sign (+) for acceleration and a minus sign (-) for deceleration.

[0036] The objective is to obtain a target IWRC value for the rolling resistance index for the test cycle considered. To this end, the method of the invention calculates, during a considered time interval IT(n_i), an effective value IWR(nl) of the rolling resistance index and, at the following time interval ITn, a time lag of anticipation Td(n) for this interval ITn is adjusted according to an error, E_IWR(nl) = IWR(nl) - IWRC, between the effective value and the target value of the rolling resistance index.

[0037] As seen in [Fig.2], the processing carried out by the C_CS software sub-module It comprises several functional blocks including an initialization block B0 and a processing block BT. The processing block BT is executed at each time interval IT. Thus, in the logic diagram of [Fig.2], sequences SQ(nl), SQn, and SQ(n+l) of the BT block are represented, which are executed respectively during the time intervals IT(ni), ITn, IT(n+i).

[0038] In the initialization block B0, the rolling speed setpoint template GA_V and the step PQ are entered (with PQ = 100 ms in this example), as well as the value of the target rolling index IWRC, a parameterizable fixed time Dopt which is an adjustable optimization parameter (initialized to 0.5 ms in this example), an initial duration D0 for the anticipation time offset Td(n) at n=0 (with Td(0) = D0 = 800 ms in this example), and minimum durations Dmin and maximum durations Dmax for the offset Td(n) (with Dmin = 500 ms and Dmax = 1300 ms in this example).

[0039] In the invention, the fixed Dopt time is configurable, allowing for optimization by modifying the convergence rate of the anticipation time lag Td of the setpoint. The Dopt time is the increment used in each time interval IT to modify, if necessary, the value of the anticipation time lag Td, by incrementing or decrementing it. Typically, this Dopt time is set only once during a standard electric-mode range test cycle of an electrified vehicle.

[0040] With reference to the SQn sequence illustrated in detail in [Fig.2], the BT processing block consumes the IWR(nl), Td(nl) and E_IWR(n-2) data provided by the SQn-1 sequence and provides as output the rolling speed setpoint CSa(n) to be applied in the time interval ITn and the IWR(n), Td(n) and E_IWR(nl) data intended for the following sequence SQn+1.

[0041] Still considering the sequence SQn, the BT block includes a functional block Fl that calculates the error E_IWR(nl) = IWR(nl) - IWRC. The calculated error E_IWR(nl) is provided to a conditional block F2, which compares it to the error E_IWR(n-2). In case of equality, that is, if E_IWR(nl) = E_IWR(n-2), the block F2 activates If the output is OK, processing continues with a functional block F3 which determines the anticipation time lag Td(n) = Td(nl). Otherwise, if the errors E_IWR(nl) and E_IWR(n-2) are different, block F2 activates a NOK output and processing continues with a conditional block F4.

[0042] Block F4 determines whether the error E_IWR(nl) is positive or negative. If the error is positive (E_IWR(nl)>0), block F4 activates an OK output and processing continues with a functional block F5 which determines the anticipation time lag Td(n) = Td(n-1)+ Dopt. Otherwise (E_IWR(nl)<0), block F4 activates a NOK output and processing continues with a functional block F6 which determines the anticipation time lag Td(n) = Td(nl)- Dopt.

[0043] Once the anticipation time lag T(n) is determined as indicated above (equal to Td(nl), Td(n-1)+ Dopt or Td(nl)- Dopt as appropriate), it is transmitted to a functional block F7.

[0044] Block F7 converts the anticipation time offset T(n), which is in milliseconds, into an integer number x(n) of anticipation time intervals (TI), by applying the equality x(n) = [Td(n) / PQ].

[0045] The number of anticipation lag time intervals x(n) given by block F7 is then processed by a functional block F8, which is responsible for integrating the duration limitations Dmin and Dmax defined for the lag Td(n). Thus, with the illustrative values ​​considered Dmin = 500 ms, Dmax = 1300 ms and PQ = 100 ms, block F8 limits the number x(n) between 5 and 13 (500 ms / 100 ms and 1300 ms / 100 ms).

[0046] The target speed to be applied, CSa(n), is then determined by a functional block F9 as being equal to the initial target CSi(n+x(n)). The initial target CSi(n+x(n)) assigned to CSa(n) is retrieved by the processing method in the template GA_V.

[0047] In sequence SQn, the process completes its processing with the execution of a function block F10. Function block F10 calculates the rolling index IWR(n) which will be used in the following sequence SQn+1. The rolling index IWR(n), as well as the time offset Td(n) and the error E_IWR(nl) are stored by the process for the following sequence SQn+1.

[0048] Illustrative readings obtained with the method of the invention are shown by CAI to CA4 curves in [Fig. 4], and are to be compared with those shown by CB1 to CB4 curves in [Fig. 5] which were obtained with the prior art technique over the same test cycle, corresponding here to a short distance. In all these readings, the sampling and quantification step (PQ) is set at 100 ms and the anticipation time lag (D0) is initially defined at 800 ms, it being understood that this time lag is dynamically adjusted in the invention, but remains fixed in the prior technique throughout the entire test cycle. In addition, the parameterizable Dopt time, applied in the invention to the anticipation time lag, is defined here as 0.2 ms.

[0049] Curves CAI and CB1 illustrate the time evolution of the rolling resistance index IWRi (in %) obtained with the method of the invention and the rolling resistance index IWRTA (in %) obtained with the prior art, respectively. Curves CA2 and CB2 illustrate the time evolution of an adjustable anticipation time lag DTA! (in ms) according to the method of the invention and a fixed anticipation time lag DTAta (in ms) according to the prior art, respectively. Curves CA3 and CB3 illustrate the time evolution of a rolling speed setpoint CS (in km / h). Curves CA4 and CB4 illustrate the time evolution of an acceleration / deceleration command CAD (in %).

[0050] As shown in curve CA2, with the method of the invention, the anticipation time lag DT Ai is optimized and varies, over this illustrative short-distance course, between approximately 800 ms and 1230 ms. The performance obtained for the IWRi index is 1.13. With the prior art technique, as shown in curve CB2, the anticipation time lag DTAta is kept fixed at 800 ms and the performance obtained for the IWRTA index is 1.22. As shown by this example, the real-time optimization of the anticipation time lag provided by the invention results in a significant improvement in the performance obtained for the IWR rolling resistance index.

[0051] The invention allows for consideration of the dynamics of the speed setpoint template during the test cycle. By enabling adaptation of the control behavior during the test cycle, the invention provides improved tracking of the speed setpoint and the achievement of a valid rolling resistance index.

[0052] The invention is not limited to the particular embodiments described herein by way of example. A person skilled in the art may, depending on the applications of the invention, make various modifications and variations falling within the scope of the invention's protection.

Claims

Demands

1. A control method implemented in a driving robot (1) managing the tracking by a vehicle (2) mounted on a roller test bench of a rolling speed setpoint (CSi) provided by a template stored in memory (GA_V), said method being of the type in which an anticipation time offset (Td) is introduced into said rolling speed setpoint (CSi, CSa), characterized in that it comprises recurring steps, carried out during a test cycle, of: a) measuring in real time a rolling index said "IWR", and b) then adjusting a current value (Td) of said anticipation time offset as a function of an error (E_IWR) between said rolling index (IWR) measured in step a) and a predetermined target rolling index (IWRC).

2. Method according to claim 1, characterized in that, in step b), said current value (Td) of said anticipation time offset is adjusted by subtracting or adding to it a parameterizable fixed time (Dopt) depending on whether said error (E_IWR) has a positive or negative sign.

3. Method according to claim 1 or 2, characterized in that, said template stored in memory (GA_V) being sampled and quantized with a predetermined step (PQ), said step a) is carried out during a first time interval (ITn.i) and said step b) is carried out during a subsequent time interval (ITn).

4. A method according to any one of claims 1 to 3, characterized in that the current value (Td) of said anticipation time lag is limited between a predetermined maximum duration (Dmax) and a predetermined minimum duration (Dmin).

5. Driving robot (1) of a vehicle (2) mounted on a roller test bench, said driving robot (1) comprising a computer (10), an actuator control device (11), an actuator assembly (12) coupled to said vehicle (2) and a digital interface (13) with said vehicle (2), characterized in that said computer (10) has a memory (MEM) storing program instructions (C_CS) for the implementation of the method according to any one of claims 1 to 4 when said program instructions (C_CS) are executed by a processor of said computer (10).

6. Assembly comprising a vehicle (2) mounted on a roller test bench and a driving robot (1) installed in said vehicle (2), characterized in that the driving robot is a driving robot (1) according to claim 5.

Images