Method for starting autonomous driving of a vehicle

Abstract

The present invention relates to a method for autonomous navigation in the case of low or zero speed maneuvers of a motor vehicle, comprising a step of correcting the orientation of the steering wheels of the vehicle when the speed of the vehicle is zero or substantially zero or below a threshold value, in particular equal to 1 km / h.

Description

[Technical field]

[0001] The invention relates to a method for autonomous navigation or autonomous control of a motor vehicle. The invention also relates to a system for autonomous navigation or autonomous control of a motor vehicle. The invention also relates to such an autonomous navigation system or to a motor vehicle equipped with hardware and / or software means implementing such an autonomous navigation method. The invention further relates to a computer program product recorded on a carrier and comprising program code instructions capable of executing the steps of the autonomous navigation method upon reading by an electronic control unit. The invention also relates to a data record carrier readable by an electronic control unit and having the program product recorded thereon. Finally, the invention relates to a signal from a data carrier comprising a computer program product.

[0002] In particular, the present invention relates to a precise method for autonomous driving in situations where the vehicle's driving speed is low, starting a journey in autonomous mode regardless of the situation surrounding the vehicle. This solution solves the problem of poor vehicle alignment when starting off or in curves. This solution significantly improves the autonomy of the vehicle. [Background technology]

[0003] An autonomous vehicle can manage its travel trajectory by itself without any action from the vehicle user, except for specifying a destination or end point of the trip. For this reason, an autonomous vehicle typically has an autonomous driving mode that allows it to be guided by an autonomous driving system.

[0004] The autonomous driving mode manages the control of the different actuators that allow the vehicle to move along a trajectory: these actuators are mainly the power unit that drives the car and the steering actuator that allows the car's steering wheels to be properly oriented.

[0005] An autonomous vehicle typically has an autonomous cruise mode that is managed by an autonomous cruise system and can be guided at a cruising speed (e.g., speeds above 30 km / h or 50 km / h). This mode is used on urban arterials and suburban highways or freeways. The autonomous cruise mode is only activated after the autonomous vehicle has reached cruising speed. Thus, the vehicle is driven by a human driver or a remote operator when the vehicle is launched or set off, when the vehicle is pulled off the road (e.g., starting from a curb), or when the vehicle is pulled over (e.g., parking at a curb), and more generally when maneuvering at low speeds, i.e., less than 10 km / h or less than 5 km / h.

[0006] On-demand mobility services are designed to transport individuals door-to-door and provide passengers with the comfort of being taken to and from their preferred locations. However, these services require autonomous vehicles that can manage autopilot with extremely precise longitudinal and lateral commands, especially during pull-over or pull-off maneuvers, to get the vehicle close enough to the curb while avoiding injury to the user or third parties or physical damage to the autonomous vehicle, other surrounding vehicles, or surrounding infrastructure. These maneuvers are complex because they require precise longitudinal and lateral positioning data at very low speeds. These requirements are not met by currently available autonomous driving systems. Autonomous driving systems are essentially dedicated to non-urban roadways and can be activated once the vehicle is in motion. However, if the autonomous driving system is activated when the vehicle is stationary or moving at low speeds, the trajectory management is insufficient, increasing the risk of injury or damage associated with the vehicle's movement.

[0007] Precise low-speed control is paramount for successful management of autonomous low-speed maneuvers, specifically those that require a small steering radius while ignoring the dynamic data and current state of the vehicle.

[0008] Shortly after activating the autonomous mode of a vehicle traveling at zero or virtually zero speed, unexpected events may cause the vehicle to be hindered, requiring a teleoperator to take control of the vehicle to solve the problem. These problems arise due to (overly) approximate estimation of the vehicle's state, which may ignore, for example, heading, wheel direction, or lateral position errors, resulting in the calculation of a trajectory that is deemed impossible (e.g., beyond the vehicle's capabilities).

[0009] In other words, from zero or substantially zero speed to cruising speed, the vehicle is not autonomous, and as a result, an in-vehicle driver or a remote operator must be able to be summoned, particularly when the vehicle arrives at a ramp or curb (pulling over) to unload and / or pick up occupants, when the vehicle departs from a ramp or curb (pulling over), or when performing complex maneuvers.

[0010] Document US9645577 discloses a guidance system designed to autonomously guide a vehicle in a constrained environment. The system generates different space-time trajectory solutions as different movement strategies so that the vehicle can choose the one that it considers optimal. Once the strategy is decided, the necessary commands to follow it are sent to the vehicle's actuators. The disadvantage of this solution is that at low speeds, it may result in the selection of a trajectory that is not feasible depending on the current state of certain actuators. Summary of the Invention

[0011] The object of the present invention is to provide an autonomous driving method that overcomes the above mentioned disadvantages and improves the driving methods known from the prior art. In particular, the present invention makes it possible to produce autonomous vehicles capable of autonomously managing complex maneuvers performed when the vehicle is traveling at low speeds.

[0012] To this end, the invention relates to a method for autonomous driving of a motor vehicle, in particular at low or zero speed, comprising a step of correcting the orientation of the steering wheels of the vehicle in conditions where the speed of the vehicle is zero or substantially zero or below a threshold value, in particular equal to 1 km / h.

[0013] The step of correcting the orientation of the steering wheels of the vehicle includes: Immediately after the step of activating the autonomous driving mode, and / or In an autonomous mode, prior to the step of driving or moving the vehicle along a steering trajectory, It is possible to do this.

[0014] The present invention also provides a method for autonomous driving of a motor vehicle, comprising the steps of: an autonomous driving mode including the execution of the method as defined above; an autonomous cruising mode for cruising along a cruising trajectory; wherein the two autonomous driving modes (autonomous pilot driving mode and autonomous cruising driving mode) are exclusive.

[0015] The operational logic of the autonomous driving mode and the autonomous cruising mode may be different, and / or in the autonomous driving mode the vehicle may run or move at a speed less than 10km / h or 5km / h.

[0016] This method is defining a reference cruise trajectory; - defining, in particular iteratively and / or by simulation, a steering trajectory that satisfies the reference cruise trajectory, in particular from an initial position of the vehicle; may also include

[0017] This method is the physical ability to operate a vehicle; and / or the safety of property and / or persons in and / or near the vehicle; The method may include verifying the steering trajectory against

[0018] This method is - determining or measuring the current position of the vehicle, in particular the current location and / or the current heading; comparing a current position of the vehicle with positions constituting a reference cruise trajectory; automatically switching from the autonomous driving mode to the autonomous cruising mode when a current position of the vehicle corresponds or is believed to correspond to one of the positions constituting the reference cruising trajectory; may also include

[0019] The invention further relates to a system for autonomous driving of motor vehicles, comprising hardware and / or software elements for carrying out a method as defined above, in particular hardware and / or software elements designed to carry out a method as defined above, and / or comprising means for carrying out a method as described above in this specification.

[0020] The invention further relates to a motor vehicle equipped with hardware and / or software elements for carrying out a method as defined above, in particular with hardware and / or software elements designed to carry out a method as defined above, and / or with a vehicle equipped with means for carrying out a method as defined above.

[0021] The invention further relates to a computer program product comprising program code instructions recorded on a carrier readable by an electronic control unit and which, when running on an electronic control unit, perform the steps of the autonomous driving method as defined above, or a computer program product capable of being downloaded from a communications network and / or recorded on a computer-readable and / or computer-executable data carrier and comprising instructions which, when executed by a computer, cause a computer to perform a method as defined above.

[0022] The invention further relates to a data record carrier having recorded thereon a computer program readable by an electronic control unit and comprising program code instructions for performing the method as defined above, or a record carrier readable by a computer and comprising instructions which, when executed by a computer, cause the computer to perform a method as defined above.

[0023] The invention further relates to a signal from a data carrier containing a computer program product as defined above.

[0024] The above objects, features and advantages of the present invention will be described in detail in the following description of one specific embodiment, given in conjunction with the accompanying drawings, without implying any limitation thereto. [Brief description of the drawings]

[0025] [Figure 1] FIG. 1 is a schematic diagram of one embodiment of an autonomous vehicle. [Diagram 2] FIG. 1 is a partial schematic diagram of an embodiment of a system for autonomous navigation of an autonomous vehicle. [Diagram 3] FIG. 2 is another partial schematic diagram of an embodiment of a system for autonomous navigation of an autonomous vehicle. [Figure 4] FIG. 13 is a diagram showing the trajectory of an autonomous vehicle with an embodiment of an autonomous driving system in a roundabout situation and the trajectory of an autonomous vehicle without an embodiment of an autonomous driving system in the same situation, from above. [Figure 5a] 1 is a graph showing steering wheel angles of an autonomous vehicle with an embodiment of an autonomous driving system in a roundabout situation and an autonomous vehicle without an embodiment of an autonomous driving system in the same situation. [Figure 5b] 1 is a graph showing the speed of an autonomous vehicle with an embodiment of an autonomous driving system in a roundabout situation and the speed of an autonomous vehicle without an embodiment of an autonomous driving system in the same situation. [Figure 5c]1 is a graph illustrating lateral error of an autonomous vehicle with an embodiment of an autonomous navigation system in a roundabout situation and a lateral error of an autonomous vehicle without an embodiment of an autonomous navigation system in the same situation. [Figure 5d] 1 is a graph illustrating heading error of an autonomous vehicle with an embodiment of an autonomous navigation system in a roundabout situation and an autonomous vehicle without an embodiment of an autonomous navigation system in the same situation. [Figure 6] FIG. 2 is a top view of the trajectory of an autonomous vehicle with an embodiment of an autonomous driving system in a curved situation and the trajectory of an autonomous vehicle without an embodiment of an autonomous driving system in the same situation. [Figure 7a] 1 is a graph showing the steering angle of an autonomous vehicle with an embodiment of an autonomous driving system in a curved situation and the steering angle of an autonomous vehicle without an embodiment of an autonomous driving system in the same situation. [Figure 7b] 1 is a graph showing the speed of an autonomous vehicle with an embodiment of an autonomous driving system in a curve situation and the speed of an autonomous vehicle without an embodiment of an autonomous driving system in the same situation. [Figure 7c] 1 is a graph illustrating lateral error of an autonomous vehicle with an embodiment of an autonomous driving system in a curve situation and an autonomous vehicle without an embodiment of an autonomous driving system in the same situation. [Figure 7d] 1 is a graph illustrating heading error of an autonomous vehicle with an embodiment of an autonomous driving system in a curve situation and heading error of an autonomous vehicle without an embodiment of an autonomous driving system in the same situation. [Figure 8] FIG. 2 is a diagram illustrating an example of a man-machine interface of an embodiment of an autonomous driving system. [Figure 9] 1 is a flowchart of one aspect of executing an autonomous driving method. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] 1 shows a schematic representation of an embodiment of a vehicle 300, in particular an automobile. The vehicle 300 may be, for example, a car, a utility vehicle, a truck, or a bus.

[0027] The vehicle 300 comprises an autonomous driving system 200. The autonomous driving system is capable of managing an autonomously piloted driving mode that can manage the autonomous movement of the vehicle at low speeds, especially in complex maneuvers such as on- or off-the-shoulder maneuvers. Advantageously, the autonomous driving system can also manage a second autonomous driving mode, an autonomous cruising driving mode, that can manage the autonomous movement of the vehicle at high speeds, especially in suburban driving on highways or freeways or on essential routes in urban environments.

[0028] The autonomous driving system 200 includes: A computer 100; A set of actuators 220; a set of elements 10, 11, 12, 13, 14, 15 providing information about the state of the vehicle and its surroundings; Equipped with.

[0029] The autonomous driving system 200 may also advantageously include a human-machine interface 210. For example, the human-machine interface may inform a user, passenger, or remote operator of the feasibility of an autonomous maneuver and / or how to correct vehicle actuator commands to adjust the trajectory.

[0030] The set of actuators includes in particular a steering actuator 221, i.e. an actuator for the orientation of the steering wheels of the vehicle 300, and a drive unit 222 for powering the vehicle 300. The vehicle drive unit may be of the combustion engine type, of the hybrid drive unit type or of the electric motor type.

[0031] The set of elements 220 that provide information about the vehicle's state and surroundings are: A lane sensor 10; An obstacle sensor 11; A sensor 12 relating to the steering or orientation of the steering wheel; A vehicle speed sensor 13; A vehicle position sensor 14; a module 15 for providing a reference trajectory; may include one or more of:

[0032] One or more of the sensors may be replaced by elements providing equivalent information (eg, replaced by an estimator).

[0033] The position sensor 14 or position estimator advantageously provides information regarding the location of the vehicle, such as GPS coordinates, and vehicle orientation or vehicle heading information, for example the orientation of the vehicle's longitudinal axis.

[0034] Therefore, the computer 100 Information from sensors mounted on the vehicle (current speed and current steering angle or current steering wheel direction angle), Information from cameras or other devices capable of detecting lane markings; Information from any perception system, including cameras, radar, ultrasonic devices, lidar devices, or combinations thereof, that detect any nearby obstacles; and / or Information from a vehicle positioning system capable of providing the current position and heading of the vehicle as well as a given reference trajectory, such as a series of points to be followed by the vehicle; is used as the primary input.

[0035] From all this information, an available driving area (this area represents the boundary that the vehicle will not cross if it autonomously starts a maneuver. If the vehicle is outside this area, it means that a collision may have occurred. This information may be provided by combining information from lane sensors or detectors 10 and obstacle sensors or detectors 11. This information can later be used in a launch or launch verification module 1 (see below) to allow or stop a maneuver); the current state of the vehicle (including the current orientation of the steering wheel, for example by measuring the current angle of the steering wheel; not only the current heading of the vehicle but also the orientation of the steering wheel plays an important role in improving the maneuverability of the vehicle); A reference trajectory (this reference trajectory is provided, for example, by an autonomous path prediction module or, if the vehicle is in assisted mode due to an unknown environment, by a teleoperator); It is possible to calculate

[0036] The computer 100 a launch or launch verification module 1 capable of determining the feasibility of a maneuver in an autonomous mode and / or determining whether the vehicle can launch or launch autonomously in sufficient safety conditions; a command simulation module 2 capable of simulating commands for controlling actuators of the vehicle, in particular commands for controlling the vehicle drive units and commands for controlling the actuators for steering the vehicle's steering wheels; a trajectory following module 3 capable of correcting actuator control commands, in particular whilst performing the maneuver in autonomous mode; Equipped with.

[0037] As shown in FIG. 2, one embodiment of the command simulation module 2 includes: 4. Vehicle modeling module; Command Modeling Module 5 and Steering modeling module 6; Includes.

[0038] As shown in FIG. 3, one embodiment of the trajectory following module 3 includes: A command adjuster 30; an initial steering command correction module 37; A position correction module 38; An autonomous cruise mode module 39; Includes.

[0039] The trajectory following module 3, and more specifically the command regulator 30, Departure or departure verification information 31; Steering command model information 32; Vehicle steering wheel current angle orientation information 33; Steering command model error information 34; Vehicle lateral position error information 35; Vehicle heading or orientation error information 36; and at an input to receive signals and / or information according to

[0040] The motor vehicle, in particular the autonomous driving system and / or the computer, comprises hardware and / or software means capable of or arranged to carry out the methods constituting the subject of the present invention.

[0041] The hardware and / or software means may comprise software modules.

[0042] Hereinafter, one execution mode of the autonomous driving method will be described with reference to Fig. 9. This execution mode can also be considered as an execution mode of a vehicle driving method or an autonomous driving system. Hereinafter, this execution mode will be described in detail as an execution mode of an autonomous driving system driving method.

[0043] The vehicle is assumed to be initially stationary after one or more occupants have met and boarded the vehicle to be taken by the vehicle to a desired destination, and the vehicle is assumed to be in an environment with obstacles and thus requiring complex maneuvers to be able to start or depart or pull off the road.

[0044] It is also assumed that the vehicle or autonomous driving system has a reference trajectory or has calculated a reference trajectory. The reference trajectory is a trajectory that allows the vehicle to move from a starting point to a destination point following or substantially following the lines of the lane. Thus, when the vehicle follows the reference trajectory, it is located in the center of the lane or substantially in the center. Such a trajectory is incompatible with vehicle conditions where the vehicle is parked near a curb, in a yard, or in a parking lot. That is, in such situations, the position (location and orientation) of the vehicle does not correspond to any of the positions defined by the reference trajectory.

[0045] In a first stage 110, the autonomous driving system operates in an autonomous driving mode to simulate vehicle states, vehicle actuator commands, and steering trajectories.

[0046] A maneuvering trajectory is a vehicle trajectory that may satisfy the reference trajectory from its initial state (e.g., near a curb, in a yard, or parked in a parking lot).

[0047] As shown in Figures 1 and 2, the command simulation module 2 is fed with various information. This allows the generation of a virtual model of the vehicle that corresponds to the current state of the vehicle. The vehicle model is generated in the same position as the real vehicle, but is assigned a minimum speed that allows the calculation of actuator commands to follow a reference trajectory, in particular the commands of the actuators orienting the steering wheels. In this vehicle model, the orientation of the steering wheels (and thus the position of the steering wheel) is constantly corrected until the moment when the trajectory following module 3 indicates that the commands determined to control the actuators allow a correct start of the maneuver or that the actuator commands have reached the physical limits of the vehicle (for example a steering wheel angle indicating a maximum right or left lock).

[0048] The execution mode steps of this first phase 110 are described in more detail below.

[0049] In a first step, the actuator command simulation module 2, in particular the vehicle modelling module 4, Current information on the orientation of the steering wheel (sensor 12) (this information allows the simulation of the vehicle trajectory if the autonomous driving system does not correct the current orientation of the steering wheel), Vehicle speed information (sensor 13) (this information may come from an on-board sensor that ensures that the vehicle is completely or substantially stationary. For example, if the vehicle is traveling at cruising speed, the autonomous driving mode is deactivated. In this case, an autonomous cruising mode may be activated, which allows the error with respect to the reference trajectory to be managed), Vehicle positioning information (sensors 13) that can be provided at a local level (for example by means of a camera) or at a global level (for example based on GPS) (which allows the current position of the vehicle and its orientation to be known, which information is naturally related to the orientation of the steering wheel of the vehicle), said reference trajectory (originating from a module 15 providing said reference trajectory) (this module may form part of a system enabling the implementation of an autonomous cruise mode using said reference trajectory, or said reference trajectory may be transmitted by a remote operator in the case of vehicle assistance); The information may include, as input, all or part of the above.

[0050] In a second step, using the reference trajectory, the vehicle modeling module 4 is used to virtually model the vehicle by imposing a minimum speed on the vehicle while ensuring that its position does not virtually move, meaning that the modeled virtual vehicle is still in the same virtual position (e.g., the longitude, latitude and heading values ​​are identical).

[0051] In a third step, the command modeling module 5 corrects the steering wheel orientation error that exists between the current orientation of the steering wheel and the orientation required to follow the reference trajectory, and as a result, the command modeling module 5 generates commands to control the vehicle steering wheel orientation actuators.

[0052] In a fourth step, the commands generated in the third step are fed to the steering modelling module 6, on the basis of which the position of the virtual vehicle can be changed in the future to ascertain the future values ​​of the vehicle's position (longitude, latitude and heading) relative to the boundaries of the driving area available in the vicinity of the vehicle. Then, finally, a check is made as to whether the generated commands result in a virtual orientation of the steering wheels that allows reaching the desired point on the reference trajectory or does not exceed the physical capabilities of the vehicle in terms of the wheel orientation angle. If this check is not OK, the method returns to the third step, In the first scenario, the final generated commands result in a virtual steering wheel orientation that allows reaching the desired point on the reference trajectory, or In the second scenario, the generated commands exceed the vehicle's physical capabilities with respect to the angular orientation of the vehicle's wheels. The third and fourth steps are repeated iteratively until

[0053] The method therefore includes the step of validating the steering trajectory against the physical steering capabilities of the vehicle.

[0054] In the first scenario, the method proceeds to a second stage 120. The last values ​​of the commands of the actuators directing the wheels of the vehicle are retained.

[0055] In the second scenario, the autonomous driving system is unable to find actuator commands that would allow it to execute the maneuver. As a result, the autonomous driving system notifies the vehicle user or teleoperator of this fact. This information may be communicated, for example, via the human-machine interface 210.

[0056] Thus, in this first stage, the method comprises: defining or receiving a reference cruise trajectory; - defining, in particular iteratively and / or by simulation, a steering trajectory that satisfies the reference cruise trajectory, in particular from an initial position of the vehicle; Includes.

[0057] In a second stage 120, the autonomous driving system tests whether it is possible to perform complex maneuvers in autonomous mode or with a sufficient level of safety for individuals and assets in the vehicle and in the external environment close to the vehicle. If possible, the autonomous driving system executes stage 130. If not possible, the autonomous driving system executes stage 140. Thus, in the second stage 120, the autonomous driving system generates a virtual trajectory for the vehicle to follow by evaluating the vehicle and the actuator commands previously determined (by the command simulation module 2). The virtual trajectory is checked by the launch or launch verification module 1 to ensure a safe launch (e.g., the vehicle's predicted future position is inside the driving area). For this purpose, the vehicle speed is used to simulate the future changes in the position of the virtual vehicle during the maneuver. This speed can be parameterized and is fixed, for example, at 2 m / s.

[0058] Advantageously, the autonomous driving system may command the human-machine interface to display and / or illustrate certain or all iterations (until it is clear whether or not a feasible maneuver exists). The autonomous driving system may preferably command the human-machine interface to display information related to the performance of the autonomous driving system (e.g., a simultaneous display of the last adopted virtual trajectory and the reference trajectory provided by module 15, or a simultaneous display of the last calculated initial orientation of the vehicle's steering wheel and the current orientation of the vehicle's steering wheel). The autonomous driving system may a safe launch indicator that indicates that actuator control has been found to autonomously execute a safe or collision-free maneuver; Autonomous Route Commencement Warning Indicator, and / or an assist indicator to indicate to the user and / or remote operator that manual / external intervention is required; Preferably, the display of the image can be commanded to the man-machine interface.

[0059] The method therefore includes a step of verifying the maneuver trajectory with respect to the safety of assets and / or individuals in and / or near the vehicle.

[0060] In stage 130, the autonomous driving system generates and executes real commands to control the vehicle's actuators to perform the complex maneuver. If the validation indicator indicates that a safe launch maneuver is possible, the autonomous driving system activates the trajectory following module 3, which receives virtual commands determined by the command simulation module 2 and enabling the execution of the maneuver according to the determined virtual trajectory. The validation indicator then stops the iterative calculations of the vehicle actuator command simulation module 2 and adopts the actuator command values ​​from the final simulation. A three-step procedure is then performed by the trajectory following module 3. The final simulated actuator commands (especially the final simulated actuator commands for steering wheel orientation) are considered as the initial commands applied to the actuators at the beginning of the execution of the maneuver. However, these commands are applied when the vehicle is stationary, i.e. at zero speed. Here, the autonomous driving system defines the initial steering angle value or steering wheel orientation that the vehicle needs to have while maintaining zero speed. In this step, the difference between the virtual steering command of the steering wheel and the actual steering of the steering wheel is minimized at the beginning of the execution of the maneuver. If this difference is below a given threshold, the autonomous driving system executes the maneuver and the vehicle can start its journey autonomously. This stage is important because it allows correction of imperfect alignment (e.g., a vehicle with lateral position error, significant heading error with respect to the virtual trajectory, and / or steering wheel misorientation) before the vehicle moves so that the maneuver can start in the correct direction. The minimum speed is fixed (for example 1km / h or 10km / h) so that the vehicle can compensate for any imperfect alignment. Angular and / or lateral error thresholds are determined and provided to define the moment when the vehicle is considered to have properly corrected its initial position and is ready to activate the autonomous cruise mode.

[0061] Such a procedure is well accepted by users and so is not surprising.

[0062] This third phase 130 execution mode step (previously described herein) is described in more detail below.

[0063] In a first step, the trajectory following module 3, in particular the command regulator 30, Launch or launch verification information 31 (this information indicates that a safe launch is possible and that the maneuver can be performed autonomously and safely); Steering command model information 32 (this information includes the final commands to the actuators for orienting the steering wheels of the vehicle, calculated in the last iteration of the third step of the first stage 110); Vehicle steering wheel current angle orientation information 33; Steering command model error information 34 (this information includes the difference between the current orientation of the vehicle's steering wheels and the orientation of the virtual vehicle's steering wheels); Vehicle lateral position error information 35 (this information includes a value of the distance separating the reference trajectory from the vehicle's center of gravity position); Vehicle heading or orientation error information 36 (this information includes the value of the angle between the vehicle's longitudinal axis and the tangent to the reference trajectory); The whole or part of the information is received via input.

[0064] All of these inputs allow regulator 30 to initiate different command actions depending on the vehicle conditions.

[0065] Thus, in the first step of the third stage 130, the initial direction command correction module 37 has the task of minimizing the steering command model error value 34. To this end, a zero or substantially zero speed is imposed on the vehicle as long as the orientation of the steering wheels of the vehicle has been corrected to achieve the orientation determined at the end of the first stage. Thus, the method is carried out in such a way that the speed of the vehicle is: zero, Virtually zero, or below a threshold value, in particular a threshold value equal to 1 km / h, The method includes the step of modifying the orientation of the steering wheels of the vehicle while the vehicle is in a neutral position.

[0066] The step of correcting the orientation of the steering wheels of the vehicle includes: Immediately after autonomous driving mode is activated, and / or In an autonomous mode, before the vehicle drives or moves along a steering trajectory, is executed.

[0067] Advantageously, regulator 30 disengages module 37 when the value of error 34 falls below a predetermined threshold.

[0068] Then, in a second step of the third phase 130, the regulator 30 activates the position correction module 38. This module 38 fixes a given speed (for example 2 m / s and / or for example identical to the speed used in the first phase 110). The vehicle then actually moves by its drive units so as to correct the position errors, in particular the orientation and lateral position errors present with respect to the points of the reference trajectory to be met. The maneuver is therefore performed in autonomous mode. In the autonomous driving mode, the vehicle preferably drives or moves at a speed below 10 km / h or below 5 km / h. The vehicle position error values ​​are preferably constantly measured and / or estimated and then compared with respective predefined thresholds (typically 30 cm for lateral position errors and 0.1 radians for orientation errors). When all the position error values ​​are below the respective predefined thresholds, the regulator 30 deactivates the module 38. The maneuver is then completed.

[0069] Next, in a third step of the third phase 130 , the regulator 30 activates the autonomous cruise mode module 39 .

[0070] Therefore, this method - determining or measuring the current position of the vehicle, in particular the current location (latitude, longitude) and / or the current heading; comparing a current position of the vehicle with positions constituting a reference trajectory; automatically switching from the autonomous driving mode to the autonomous cruising driving mode when a current position of the vehicle corresponds or is believed to correspond to one of the positions constituting the reference trajectory; Includes.

[0071] In this manner, the autonomous driving method allows for an automatic and / or user-initiated transition from an autonomously piloted driving mode to an autonomous cruising driving mode. an autonomous driving mode, which includes, in particular, the execution of a first step of a third phase 130; Autonomous cruising mode; The two autonomous driving modes (autonomous pilot driving mode and autonomous cruising driving mode) are exclusive.

[0072] The calculation logic of the autonomous driving mode and the autonomous cruising mode is different; The vehicle is driven or moved at a speed below 10km / h or 5km / h in autonomous driving mode; and / or that in the autonomous cruise mode, the vehicle drives or moves normally (without hindrance, in particular without hindrance related to traffic) at a speed equal to or substantially equal to the permitted speed limit of that lane; is preferred.

[0073] In step 150, the autonomous driving system tests whether the vehicle is on the trajectory defined by the autonomous cruise driving mode, and if so, the autonomous driving system exits the autonomous pilot driving mode and automatically switches to the autonomous cruise driving mode.

[0074] In step 140, the autonomous driving system is unable to find actuator commands that allow the execution of the maneuver to a satisfactory standard. As a result, the autonomous driving system informs the vehicle user or operator of this fact. This information is transmitted, for example, via a man-machine interface. Control of the vehicle's actuators is then performed by the vehicle user or operator, in particular a teleoperator, in order to realize the maneuver.

[0075] Once the maneuver has been performed and the vehicle is moving, an autonomous cruise mode can be activated in step 160. The autonomous cruise mode is activated by a user or operator action.

[0076] This allows the autonomous driving system to recognize and account for misalignments between the vehicle's position, actuator states, and the desired trajectory before beginning to execute a maneuver. The autonomous driving system detects and corrects dangerous situations when the vehicle launches with a state inconsistent with the trajectory by commanding and modifying the vehicle's response (as needed) when in autonomous mode and / or at low speeds. The autonomous driving system continuously monitors the vehicle's state, determines its feasibility by comparison against a reference trajectory, and modifies the actuator control commands accordingly to improve trajectory following and reduce risk.

[0077] The embodiment of the autonomous driving system described above has been coded and integrated into the Applicant's own vehicle (the ZOE® Robot Taxi) and tested with the following results:

[0078] The system was tested on tracks with different initial vehicle conditions and orientations. The results were compared against results obtained with the same vehicle without implementing the invention. These results show a significant improvement in the autonomous capabilities of the vehicle.

[0079] Figure 4 shows a first situation in which the initial situation of the vehicle with respect to the reference trajectory is characterized by a slight alignment error on the occupied lane (right lane) and a maximum right lock of the steering wheel near the roundabout entrance. Also shown are the desired trajectory TR1 and the road boundary L1, respectively. Time is shown according to the same representation from zero (starting point) to 25 seconds. Curve B represents the trajectory of the vehicle obtained in the autonomous driving mode without application of the present invention, and curve A represents the trajectory of the vehicle obtained in the autonomous driving mode with the autonomous driving system described above in this specification.

[0080] Note that curve B immediately deviates from the lane. The trajectory remains out of the lane for a long time and then returns to the desired reference trajectory (a trajectory that makes a U-turn around the roundabout). Curve A, on the other hand, stays in the lane. In particular, the solution according to the invention orients the steering wheels of the vehicle before anything else before the vehicle moves. This allows the vehicle to correctly enter the roundabout from the same initial setting.

[0081] It is important to note that the excellent stability capabilities of the autonomous driving system ensure that the vehicle never enters an unstable situation again. The large lateral position and orientation errors are due to the autonomous driving mode being initialized with non-optimal vehicle and associated actuator data.

[0082] Figures 5a, 5b, 5c and 5d show the performance of the vehicle in the two tests described herein above.

[0083] Figure 5a shows the evolution of the steering angle from an initial value of -430° during the two tests. Note that the steering angle A1 converges quickly (around 7 seconds) to the unsaturated value when the vehicle is stationary and only the steering is moving. Note, however, that without the system according to the invention (curve B1), the vehicle needs to correct the steering further to achieve clear commands and accurately follow the reference trajectory.

[0084] FIG. 5b shows the speed evolution in the two tests. The speed signal B2 is non-zero even before the regulator starts to correct the steering angle, and follows the speed given by the navigation system. In curve A2, the steering wheel reaches the correct angle and then starts to move due to the system according to the invention. That is, according to the method described above in this specification, the initial steering command correction module 37 is first activated, and then the conditions for activating the other modules are checked. In this example, the navigation speed is immediately exceeded (position correction module 38 is ignored) from the moment when the conditions for activating the autonomous cruise mode module 39 are obtained (for example, from the moment when the lateral error is below 0.3 m and the heading or orientation error is below 5°).

[0085] Figure 5c shows the evolution of the lateral error in the two tests. With the invention implemented, even in difficult situations such as a roundabout, the vehicle (curve A3) achieves a lateral error of less than 0.25 m, while without the invention (curve B3), the vehicle exceeds a lateral error of 2.8 m relative to the roadway. Similar results for the heading error are shown in Figure 5d, where with the invention (curve A4), the vehicle maintains a heading error of less than 0.2 radians, while without the invention (curve B4) this is not the case.

[0086] The second test was taken in a curved situation. The vehicle start conditions for the two tests are the same, with a heading error of around 40° relative to the desired reference trajectory and a steering angle of maximum left lock (around 480°). Figure 6 shows the trajectories adopted by the vehicles in both cases (trajectory C for the vehicle equipped with the autonomous driving system described herein above and trajectory D for the vehicle without the invention), the lane boundaries L2, and the desired reference trajectory TR2 for each lane.

[0087] In this case, both trajectories are shown, but only trajectory C is possible without crossing the boundary (in particular, without occupying the lane of oncoming traffic). Trajectory C is performed by a vehicle equipped to implement the invention, where it converges quickly towards the desired reference trajectory without crossing the oncoming lane, transitioning to normal behavior around 11 seconds. In contrast, trajectory D represents a vehicle not implementing the invention, which stabilizes around 15 seconds after starting, clearly crossing the other lane, in other words, deviating from the fixed boundary.

[0088] Figures 7a, 7b, 7c and 7d show the performance of the vehicle in the two tests described herein above.

[0089] Figure 7a shows the change in steering angle from the initial value in the two tests, and it is clear that the vehicle implementing the invention and following curve C1 converges more quickly towards unsaturated behavior than the vehicle not implementing the invention and following curve D1.

[0090] Fig. 7b shows the change in speed in the two tests. The speed curve C2 of the vehicle implementing the invention follows the commands given by the trajectory following module 3: first a zero speed is applied to converge to the correct steering angle, then a speed of 2 m / s is applied, and finally, at a third location, the autonomous cruising mode 39 is activated if the errors in the orientation and lateral position are below a predefined threshold. In contrast, the speed curve D2 rises quickly to 2 m / s, which does not leave enough time for the vehicle to converge to a stable state.

[0091] Figure 7c shows the evolution of the lateral position error during the two tests: according to curve C3 for the vehicle implementing the invention, an initial error of just under 2m at initialization is corrected and the error does not exceed 0.2m thereafter, whereas in curve D3 for the vehicle not implementing the invention, the error reaches around 3m.

[0092] Finally, Figure 7d shows the evolution of the heading error in the two tests: in curve C4 for the vehicle implementing the invention, the initial error converges quickly, whereas in curve D4 for the vehicle not implementing the invention, the error increases and then converges towards the vehicle's normal behavior.

[0093] An overview of an example of a human-machine interface 210 is shown in Figure 8. This interface can inform the occupant or a teleoperator about the feasibility of a maneuver trajectory and the response of the vehicle with corrections to adapt to a given trajectory.

[0094] On the interface 210, the trajectory that the vehicle will follow at the initial position and initial steering angle (desired trajectory TR' and trajectory A' shown on the map including the driving area) is compared with the trajectory B' that the vehicle will follow without implementing the present invention. As shown, the virtual vehicle corrects the orientation of the steering wheel of the vehicle, and then the vehicle moves. The angular position of the virtual steering wheel can be shown on the interface 210, as well as the current angular position of the steering wheel.

[0095] If the trajectory confirmed by the launch verification module 1 is feasible, an indication 211, such as a lamp 211, may be displayed on the interface.

[0096] If the trajectory confirmed by the launch verification module 1 is not feasible, an indication 212, such as a lamp 212, may be displayed on the interface.

[0097] If no possible trajectory is found after performing multiple tests, an indication 213, such as a light 213, may be displayed on the interface.

[0098] The interface can therefore show how the vehicle attempts to autonomously manage complex maneuvering situations.

[0099] The advantages of the solution described herein include an improved capability of vehicles, in particular robotaxi type vehicles, which have difficulties starting in curved zones or when there are large errors in the lateral position or orientation with respect to the reference trajectory provided by the autonomous cruise system, and the solution described above allows the management of on-coming, off-coming or remote control maneuvers, which represent maneuvers that are complex to manage by an autonomous vehicle.

Claims

1. A method for autonomous navigation of an autonomous vehicle (300), comprising: an autonomous driving mode comprising a step of correcting the orientation of the steering wheels of the vehicle when the speed of the vehicle is zero or substantially zero or below a threshold value, in particular a threshold value equal to 1 km / h; an autonomous cruising mode for cruising along a reference trajectory; wherein the method comprises: A step (120) of verifying whether the steering trajectory is capable of traveling in the autonomous cruising mode; and depending on the result of the verifying step, transitioning the automated vehicle to a state (130) configured to automatically switch (150) from the autonomous driving mode to the autonomous cruise driving mode or transitioning to a manual mode (140) in which the automated vehicle can be manually switched (160) to the autonomous cruise driving mode.

2. 2. The method of claim 1, wherein the autonomous piloting driving mode and the autonomous cruise driving mode are exclusive.

3. In the autonomous driving mode, the vehicle drives or moves at a speed of less than 10 km / h or less than 5 km / h; 3. The method according to claim 1 or 2, characterized in that

4. The verifying step comprises: defining the reference trajectory; defining a steering trajectory that satisfies the reference trajectory from an initial position of the motor vehicle; 4. The method according to claim 1, further comprising:

5. 4. The method according to claim 1, wherein the step of verifying is performed such that the predicted future position of the motor vehicle is located within a driving area.

6. The automated switching (150) of the automated vehicle from the autonomous driving mode to the autonomous cruising driving mode comprises: - determining or measuring a current position, in particular a current location and / or a current heading, of said motor vehicle; comparing the current position of the motor vehicle with positions constituting a reference trajectory; automatically switching from the autonomous driving mode to the autonomous cruising driving mode when the current position of the automated vehicle corresponds or is believed to correspond to one of the positions that constitute the reference trajectory; The method of claim 4, comprising:

7. A system (200) for autonomous navigation of an autonomous vehicle (300), comprising hardware and / or software elements (100, 210, 220) for carrying out the method according to any one of claims 1 to 6.

8. A motor vehicle (300) comprising hardware and / or software elements (200, 100, 210, 220) for executing the method according to any one of claims 1 to 6.

9. A computer program product comprising program code instructions recorded on a carrier readable by an electronic control unit (100) and which, when operating on said electronic control unit (100), executes the steps of the autonomous driving method according to any one of claims 1 to 6.

10. A data record carrier having recorded thereon a computer program readable by an electronic control unit (100) and comprising program code instructions for carrying out the method according to any one of claims 1 to 6.

Images