Method, device and equipment for avoiding of automatic driving vehicle and computer storage medium

Abstract

The present disclosure provides an automatic driving vehicle avoidance method and device, electronic equipment and computer storage medium, relates to the field of artificial intelligence, in particular to the field of automatic driving and intelligent traffic technology. The specific implementation scheme is: generating an avoidance decision of a target vehicle to avoid an obstacle according to a first prediction trajectory of the obstacle and an original control trajectory of the target vehicle; generating at least one first planning trajectory of the target vehicle based on current driving information of the target vehicle; the first planning trajectory is a trajectory that meets a preset speed change rule and conflicts with the avoidance decision; performing a failure operation on the first planning trajectory to reject controlling the target vehicle according to a control operation corresponding to the first planning trajectory within an avoidance range of the target vehicle to the obstacle. The embodiment of the present disclosure can generate a stable avoidance strategy, and improve the riding experience and safety.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, and in particular to the fields of autonomous driving and intelligent transportation technology. Background Technology

[0002] With the development of computer technology, computers are becoming increasingly integrated into all aspects of people's lives, including clothing, food, housing, and transportation. For example, in scenarios such as intelligent transportation, autonomous driving, and driverless vehicles, computer models can be used to control vehicles and traffic facilities, assisting or leading the generation of control suggestions or commands based on specific situations, thereby reducing manual operation.

[0003] In scenarios such as autonomous driving, driverless driving, or intelligent transportation, there are situations where vehicles encounter each other and need to determine whether to overtake or avoid. How to enable driverless vehicles to safely engage with obstacles and make stable overtaking decisions is a key issue in driverless technology and a technical challenge in this field. Summary of the Invention

[0004] This disclosure provides a method, device, electronic device, and computer storage medium for obstacle avoidance in autonomous vehicles.

[0005] According to one aspect of this disclosure, a method for obstacle avoidance by an autonomous vehicle is provided, comprising:

[0006] Based on the original control trajectory of the target vehicle and the first predicted trajectory of the obstacle, generate an avoidance decision indicating that the target vehicle should avoid the obstacle;

[0007] Based on the current driving information of the target vehicle, at least one first planned trajectory of the target vehicle is generated; the first planned trajectory is a trajectory that conforms to the preset speed change rules and conflicts with the avoidance decision.

[0008] A failure operation is executed on the first planned trajectory to refuse to control the target vehicle according to the control operation corresponding to the first planned trajectory within the target vehicle's obstacle avoidance range.

[0009] According to another aspect of this disclosure, a collision avoidance device for an autonomous vehicle is provided, comprising:

[0010] The obstacle avoidance decision module is used to generate an obstacle avoidance decision based on the original control trajectory of the target vehicle and the first predicted trajectory of the obstacle.

[0011] The first planning trajectory module is used to generate at least one first planning trajectory for the target vehicle based on the target vehicle's current driving information; the first planning trajectory is a trajectory that conforms to preset speed change rules and conflicts with the avoidance decision.

[0012] Failure Operation Execution Module: Used to execute failure operations on the first planned trajectory, so as to refuse to control the target vehicle according to the control operation corresponding to the first planned trajectory within the obstacle avoidance range of the target vehicle.

[0013] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0014] At least one processor; and

[0015] The memory is communicatively connected to the at least one processor; wherein,

[0016] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the methods in any embodiment of this disclosure.

[0017] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing a computer to perform the methods of any embodiment of this disclosure.

[0018] According to another aspect of this disclosure, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the methods in any embodiment of this disclosure.

[0019] According to the technology disclosed herein, it is possible to determine at least one first planned trajectory that the target vehicle may execute but does not conform to the avoidance decision based on the original control trajectory of the target vehicle and the first predicted trajectory of the obstacle, and to perform a failure operation on the first planned trajectory, thereby avoiding the situation where the vehicle travels along the planned trajectory without avoiding the obstacle after generating an avoidance decision to perform an avoidance operation, which would lead to a scenario that requires emergency braking and reduce the riding experience of passengers in the target vehicle.

[0020] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0021] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0022] Figure 1 This is a schematic flowchart of an obstacle avoidance method for an autonomous vehicle according to an embodiment of the present disclosure;

[0023] Figure 2 This is a schematic flowchart of an obstacle avoidance method for an autonomous vehicle according to another embodiment of the present disclosure;

[0024] Figure 3This is a schematic flowchart of an obstacle avoidance method for an autonomous vehicle according to yet another embodiment of the present disclosure;

[0025] Figure 4 This is a schematic diagram of an obstacle avoidance method for an autonomous vehicle according to an example of this disclosure;

[0026] Figure 5A This is a schematic diagram of trajectory planning based on an example of this disclosure;

[0027] Figure 5B This is a schematic diagram of the velocity-time curve according to an example of this disclosure;

[0028] Figure 6 This is a schematic diagram of trajectory planning based on another example of this disclosure;

[0029] Figure 7 This is a schematic diagram of an obstacle avoidance device for an autonomous vehicle according to an embodiment of the present disclosure;

[0030] Figure 8 This is a schematic diagram of an obstacle avoidance device for an autonomous vehicle according to another embodiment of the present disclosure;

[0031] Figure 9 This is a schematic diagram of an obstacle avoidance device for an autonomous vehicle according to yet another embodiment of the present disclosure;

[0032] Figure 10 This is a schematic diagram of an obstacle avoidance device for an autonomous vehicle according to yet another embodiment of the present disclosure;

[0033] Figure 11 This is a schematic diagram of an obstacle avoidance device for an autonomous vehicle according to yet another embodiment of the present disclosure;

[0034] Figure 12 This is a schematic diagram of an obstacle avoidance device for an autonomous vehicle according to yet another embodiment of the present disclosure;

[0035] Figure 13 This is a block diagram of an electronic device used to implement the obstacle avoidance method for an autonomous vehicle according to embodiments of the present disclosure. Detailed Implementation

[0036] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0037] This disclosure provides a method for obstacle avoidance handling by an autonomous vehicle. Figure 1This is a flowchart illustrating an obstacle avoidance method for an autonomous vehicle according to an embodiment of the present disclosure. This method can be applied to an electronic device capable of executing instructions via a front-end or terminal. For example, when deployed on a terminal (including an in-vehicle terminal), server, or other processing device, the device can perform steps such as acquiring target information and determining stability. The terminal can be a user equipment (UE), mobile device, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, in-vehicle device, wearable device, etc. In some possible implementations, the method can also be implemented by a processor calling computer-readable instructions stored in memory. Figure 1 As shown, the avoidance methods for autonomous vehicles include:

[0038] Step S11: Based on the original control trajectory of the target vehicle and the first predicted trajectory of the obstacle, generate an avoidance decision indicating that the target vehicle should avoid the obstacle;

[0039] Step S12: Based on the current driving information of the target vehicle, generate at least one first planned trajectory for the target vehicle; the first planned trajectory is a trajectory that conforms to the preset speed change rules and conflicts with the avoidance decision.

[0040] Step S13: Perform a failure operation on the first planned trajectory to refuse to control the target vehicle according to the control operation corresponding to the first planned trajectory within the target vehicle's obstacle avoidance range.

[0041] In the specific implementation method, Figure 1 The obstacle avoidance method for autonomous vehicles provided in other embodiments of this disclosure can be executed on the vehicle side, specifically on the target vehicle side. Steps S11-S13 can be started when an obstacle appears within the detectable range of the target vehicle, that is, each time the target vehicle detects an obstacle.

[0042] In this embodiment, the original control trajectory of the target vehicle can be the original trajectory of the target vehicle, which can be generated before the target vehicle departs for the destination, or when the target vehicle starts the autonomous driving or driverless driving mode, or it can be the control trajectory that the target vehicle plans to execute before detecting an obstacle.

[0043] In another possible implementation, the target vehicle's original control trajectory can be the trajectory that the target vehicle executes by default before making adjustments according to the avoidance decision, that is, the trajectory being executed when an obstacle is detected.

[0044] The first predicted trajectory of an obstacle can be a trajectory generated over a future period based on information about the detected obstacle. Alternatively, it can be a trajectory generated over a future period based on information about the detected obstacle and current environmental information. Current environmental information can include road surface information, congestion information, traffic flow information, and weather information. Obstacle information can include the obstacle's real-time speed information, as well as its acceleration information, volume information, and attribute information.

[0045] Avoidance decisions used to instruct a target vehicle to avoid an obstacle can include speed control decisions, acceleration control decisions, and direction control decisions for the target vehicle. They can also include definitive decisions regarding obstacle avoidance; for example, after a series of judgments, the target vehicle generates an avoidance decision to provide a positive indication of whether or not to avoid the obstacle.

[0046] The target vehicle's current driving information can include its current speed, current position, current acceleration, and other information.

[0047] Based on the current driving information of the target vehicle, at least one first planned trajectory of the target vehicle is generated. Alternatively, based on the current driving information of the target vehicle, at least one possible trajectory of the target vehicle is generated, and at least one first planned trajectory is determined from the at least one possible trajectory.

[0048] In this embodiment, at least one first planning trajectory can also be determined by exhaustive search. That is, according to the preset speed change rules of the target vehicle, all the planned trajectories that conflict with the avoidance decision among all possible trajectories are taken as the first planning trajectory.

[0049] The preset speed change rules can be used to control the speed of a target vehicle, including the speed change range, speed change step size range, and speed range. The preset speed change rules can be determined based on the specific driving location and road segment. For example, if the speed limit on road A is X, the speed change range can be 0-X. Similarly, if road B is currently congested, the target vehicle's speed can be set to not exceed the average speed of traffic on road B.

[0050] In one possible implementation, the preset speed change rules can be determined based on the target vehicle's built-in rules, its own performance, and the speed limit information of the current road segment. For example, if the target vehicle's built-in speed range is the Y range and its built-in acceleration range is the Z range, then the preset speed change rules are determined based on the Y range and the Z range.

[0051] In another possible implementation, the preset speed change rule can be the speed change rule under non-emergency braking conditions. This avoids a poor riding experience for passengers in the target vehicle due to sudden braking, achieving a smooth stop.

[0052] In this embodiment, the first planned trajectory is a trajectory that conforms to the preset speed change rules and conflicts with the avoidance decision. That is, the first planned trajectory is generated according to the preset speed change rules. For the target vehicle, the first planned trajectory can be executed under the current speed and position, but at the same time, it will collide with the obstacle or fail to avoid the obstacle.

[0053] In one possible implementation, when an avoidance decision is made, any planned trajectory that causes the target vehicle to accelerate within the avoidance range can be considered the first planned trajectory that conflicts with the avoidance decision. This allows for a certain speed reduction transition phase when a stopping operation is finally required, providing passengers with a smoother ride.

[0054] In another possible implementation, trajectories that do not conform to driving habits or driving regulations can also be determined as the first planned trajectory.

[0055] Performing a failure operation on the first planned trajectory can include either a temporary deletion operation or a direct deletion operation. A temporary deletion operation can be a deletion operation that invalidates the planned trajectory within the time frame during which the target vehicle may collide with the obstacle.

[0056] Performing a failure operation on the first planned trajectory may also include performing a marking operation on the first planned trajectory, so that the first planned trajectory fails within the time range required to fail.

[0057] The avoidance range of the target vehicle can be either the time range or the distance range within which the target vehicle may collide with the obstacle. It can be determined based on the target vehicle's obstacle detection information, or it can be defined as the time range or distance range within which the target vehicle can detect the obstacle. Therefore, the avoidance range can be generated based on the target vehicle's original control trajectory and the obstacle's first predicted trajectory.

[0058] In one possible implementation, the avoidance range can also be a default distance range or time range, in which case the avoidance range can be determined based on the target vehicle's current position and the default distance range. In this case, if there is still a risk of collision between the obstacle and the target vehicle after the target vehicle has passed the avoidance range, steps S11-S13 can be executed again.

[0059] In this embodiment, based on the original control trajectory of the target vehicle and the first predicted trajectory of the obstacle, at least one first planned trajectory that the target vehicle may execute but does not conform to the avoidance decision can be determined. The first planned trajectory is then invalidated, thereby avoiding the situation where the vehicle drives along the planned trajectory that does not avoid the obstacle after an avoidance decision is generated and an avoidance operation is performed, which would lead to a scenario that requires emergency braking and reduce the riding experience of passengers in the target vehicle.

[0060] In one implementation, the first predicted trajectory includes multiple paths, and the avoidance method for the autonomous vehicle further includes:

[0061] At a first set number of consecutive planning moments, acquire obstacle speed information corresponding to each planning moment;

[0062] Based on the obstacle speed information corresponding to each planning time, the first predicted trajectory corresponding to each planning time is generated.

[0063] In this embodiment, a first predicted trajectory can be generated for each detected obstacle at each planning time. At each planning time, velocity information of the obstacle can be obtained through real-time velocity detection, specifically including linear velocity information and velocity direction information.

[0064] The first set quantity can be any integer, such as 1-20.

[0065] In one possible implementation, the first set quantity can be a specific positive integer, or an integer range consisting of positive integers, such as [1, 20]. This means that, based on the current driving needs of the target vehicle, the specific situation of the obstacles, and other information, any data between 1 and 20 can be selected as the number of planning moments, and then the corresponding first predicted trajectory is generated.

[0066] In this embodiment, the planning time can be determined according to a set planning interval. The planning interval can also be determined based on road conditions and the specific driving status of the target vehicle. For example, when the road is winding and difficult to navigate, or when the vehicle speed is high in urban areas, the planning interval can be shortened.

[0067] In this embodiment, acquiring the corresponding obstacle speed information at multiple consecutive planning moments helps generate accurate avoidance decisions.

[0068] In one implementation, based on the target vehicle's original control trajectory and the obstacle's first predicted trajectory, an avoidance decision indicating that the target vehicle should avoid the obstacle is generated, including:

[0069] When there is a risk of collision between a first set number of predicted trajectories and the original control trajectory, an avoidance decision is generated.

[0070] In this embodiment, a decision to avoid collisions is generated when a first predetermined number of first predicted trajectories pose a risk of collision with the original control trajectory.

[0071] In this embodiment, the avoidance decision can represent an affirmative determination on whether to avoid the target vehicle, determining that the target vehicle should avoid the target vehicle.

[0072] In this embodiment, by setting a first predicted trajectory at a certain number of planning times and setting conditions for whether to determine avoidance, the accuracy of avoidance decisions can be improved.

[0073] In one implementation, the obstacle avoidance method for autonomous vehicles further includes:

[0074] In the event that there is a risk of collision between a second set number of first predicted trajectories and the original control trajectory, the deceleration time period is determined based on the planning time corresponding to the second set number of first predicted trajectories.

[0075] Generate control commands to control the target vehicle to perform deceleration operations during the deceleration period.

[0076] In this embodiment, the second set quantity can be one or more.

[0077] When the second set number is 1, the consecutive second set number of first predicted trajectories can represent the first predicted trajectories corresponding to adjacent planning times before and after the first predicted trajectory with collision risk, which do not have collision risk with the original control trajectory.

[0078] In one possible implementation, the second set quantity is less than the first set quantity.

[0079] The situation where there is a risk of collision between a second set number of consecutive first predicted trajectories and the original control trajectory can refer to the situation where there is no risk of collision between the first predicted trajectory corresponding to the adjacent planning time before and after the second set number of consecutive first predicted trajectories and the source control trajectory.

[0080] The deceleration time period is determined based on the planning time corresponding to the second set number of first predicted trajectories. Alternatively, the deceleration time period can be determined based on the planning time corresponding to the last one of the second set number of first predicted trajectories.

[0081] For example, based on the planning time corresponding to the last first predicted trajectory in the second set number of first predicted trajectories, a third set number of planning times can be added to obtain a time period, which can be used as the deceleration time period.

[0082] In this embodiment, the deceleration time period may also include multiple control moments. At the first control moment of the deceleration time period, an initial deceleration command is generated. At each of the remaining control moments, the deceleration command is refreshed. Thus, the specific parameters of the deceleration command corresponding to each control moment may be the same or different.

[0083] The control command to control the target vehicle to perform deceleration operation during the deceleration period can be generated when the deceleration period arrives, or the deceleration control command and the corresponding execution time can be generated in advance, with the execution time within the deceleration period.

[0084] In this embodiment, when the number of first predicted trajectories with collision risk is not less than a first set number and exceeds a second set number, a control command is generated to control the target vehicle to decelerate during the deceleration period, so as to avoid the poor riding experience caused to passengers in the vehicle by sudden and sharp deceleration when there is a need to avoid collision.

[0085] In one implementation, such as Figure 2 As shown, the current driving information includes current speed, current acceleration, and current position; based on the target vehicle's current driving information, at least one first planned trajectory for the target vehicle is generated, including:

[0086] Step S21: Based on the current speed, current acceleration, and preset speed change rules, generate at least one set of planned speed and acceleration values ​​for the target vehicle; the planned speed and acceleration values ​​include the speed and acceleration corresponding to the planning time.

[0087] Step S22: Generate at least one second planning trajectory based on at least one set of velocity and acceleration planning values;

[0088] Step S23: From at least one second planning trajectory, select the planning trajectory that conflicts with the avoidance decision as the first planning trajectory.

[0089] In this embodiment, the preset speed change rules may include preset speed value addition and subtraction rules, such as the magnitude of addition and subtraction, and the maximum range of addition and subtraction. They may also include preset acceleration value addition and subtraction rules, such as the magnitude of acceleration addition and subtraction, and the maximum range of acceleration addition and subtraction.

[0090] The velocity and acceleration corresponding to at least one planning time point can be represented as follows: each time point in at least one planning time point corresponds to a velocity and an acceleration. For example, at the first planning time point, the velocity is V1 and the acceleration is A1; at the second planning time point, the velocity is V2 and the acceleration is A2, and so on.

[0091] For each set of planned velocity and acceleration values, at least one position can be generated at the corresponding planning time by combining the target vehicle's current position and velocity. As the number of planning time points increases, the possible positions of the target vehicle also increases. For each planning time point, there may be more than two sets of planned velocity and acceleration values. For example, the preset velocity change rules include: the velocity change range can be [-v, +v], the acceleration change range can be [-a, +a], and the acceleration change step size can be Δa. At the first planning time point starting from the current time point, if the velocity and acceleration do not exceed the preset range, based on the current acceleration value 'a', there can be two accelerations: a+Δa and a-Δa, thus there may be two velocities at the first planning time point. Similarly, at subsequent planning time points, the possibilities for the corresponding velocity values ​​increase.

[0092] In this embodiment, generating at least one set of speed and acceleration planning values ​​for the target vehicle based on the current speed, current acceleration, and preset speed change rules may include: generating at least one set of speed and acceleration planning values ​​for the target vehicle corresponding to the planning time based on the current speed, current acceleration, and preset speed change rules.

[0093] In another possible implementation, the avoidance range can be determined first, and based on the avoidance range, multiple avoidance times can be determined, each of which is a time within the avoidance range. For each avoidance time, at least two sets of planned velocity and acceleration values ​​can be generated.

[0094] In another possible implementation, when determining the first planned trajectory, the number of planning moments can be determined according to the default value.

[0095] In another possible implementation, an exhaustive approach can be used to generate all possible second-planned trajectories that the target vehicle might execute based on its current speed and position information, as well as preset speed change rules. From the third trajectory, all planned trajectories that conflict with the avoidance decision are selected as the first-planned trajectory.

[0096] In this embodiment, a first planned trajectory can be determined, thereby enabling the subsequent execution of a failure operation on the first planned trajectory to prevent the target vehicle from traveling along the first planned trajectory and colliding with obstacles.

[0097] In one implementation, generating at least one second planned trajectory based on at least one set of velocity and acceleration planning values ​​includes:

[0098] In the velocity-time coordinate system, at least one time-velocity curve is generated based on at least one set of velocity and acceleration planning values ​​and at least one future planning moment;

[0099] Integrate at least one time velocity curve to obtain at least one planning position corresponding to at least one future planning moment;

[0100] A second planned trajectory is generated based on the target vehicle's current location and at least one planned location.

[0101] In this embodiment, by performing integral calculations on the time-velocity curves, the second planned trajectory corresponding to each curve can be quickly determined.

[0102] In one implementation, such as Figure 3 As shown, the avoidance decision is used to indicate that the target vehicle avoids the obstacle by stopping. The avoidance methods for autonomous vehicles also include:

[0103] Step S31: Generate at least one third planning trajectory based on the first planning trajectory and the avoidance decision; the third planning trajectory is the trajectory in the first planning trajectory that does not conflict with the avoidance decision;

[0104] Step S32: Determine the parking area based on the avoidance decision and the first predicted trajectory;

[0105] Step S33: In the third planned trajectory, set a parking control command so that the target vehicle stops before reaching the parking area.

[0106] In this embodiment, setting parking control commands in the third planned trajectory can include setting parking control commands in all third planned trajectories.

[0107] In this embodiment, when the avoidance decision is used to indicate that the target vehicle will avoid the obstacle by stopping, no acceleration operation is performed within the avoidance range in all third-planned trajectories.

[0108] In this embodiment, by setting parking control commands in the third planning trajectory, stable avoidance behavior can be generated, thus avoiding poor riding experience and increased collision risk caused by unstable avoidance.

[0109] In one example disclosed herein, the obstacle avoidance method for an autonomous vehicle includes, as follows: Figure 4 The steps shown are as follows:

[0110] Step S41: Scene Recognition. The main vehicle (equivalent to the target vehicle in the aforementioned embodiment) perception module provides information such as the location, speed, and acceleration of the obstacle vehicle; the main vehicle prediction module outputs the trajectory of the obstacle vehicle over a future period of time.

[0111] Step S42: Vehicle speed planning. The possible speed planning trajectory of the vehicle over a future period is given through sampling and other methods, equivalent to the second planning trajectory in the aforementioned embodiment.

[0112] Step S43: Pruning. Unreasonable velocity planning trajectories are pruned, where the unreasonable velocity planning trajectory is the first planning trajectory described in the aforementioned embodiment.

[0113] In specific implementation methods, unreasonable speed planning can include: speed planning trajectories that do not conform to human driving habits, speed planning trajectories that do not conform to road regulations, etc.

[0114] Step S44: Set yielding conditions based on the prediction results. The yielding conditions can be equivalent to the avoidance decision in the aforementioned embodiments. If, based on the prediction results of the first predicted trajectory of the obstacle, the obstacle may arrive at a certain position ahead at the same time as the driver vehicle, posing a collision risk, then the driver vehicle should yield. Simultaneously, record the obstacle information and the yielding decision.

[0115] Step S45: Perform deceleration. If the current frame (each frame is equivalent to the planning time in the aforementioned embodiment) results in no collision risk between the calculated obstacle and the main vehicle due to prediction line jitter, but the previous frame meets the yielding conditions and the obstacle's speed and direction are still trending towards cutting in, then a yielding decision is made based on this, by retaining 8 historical prediction lines of the intersecting frames, so that the main vehicle continues to decelerate and yield.

[0116] Step S46: Determine whether to yield based on the set threshold. If the yielding condition is triggered by obstacles in N consecutive frames, a Stop sign is set before the intersection to make the main vehicle stop and yield.

[0117] Step S47: The yielding strategy ends when the obstacle no longer overlaps with the predicted trajectory line of the main vehicle.

[0118] In another example disclosed herein, the possible speed planning trajectory of the main vehicle over a future period is given through sampling or other methods, such as... Figure 5A As shown, it may include:

[0119] Step S51: Obtain the current speed v0 of the main vehicle.

[0120] Step S52: Determine the pre-set fixed acceleration range and acceleration sampling resolution. For example, if the acceleration range is -2 m / s² to 2 m / s², and the acceleration sampling resolution is 0.1 m / s², the speed that the vehicle can reach at time t can be obtained using the formula vt = v0 + at. The acceleration range is generally set to the maximum and minimum acceleration that the vehicle can take, and is related to the performance of the target vehicle or the main vehicle, as well as the driving requirements of the target vehicle or the main vehicle. In this example, the acceleration sampling resolution can be the increase in acceleration, and the acceleration range and acceleration sampling resolution can be the speed change rules.

[0121] Step S53: Based on the current speed v0, acceleration range and acceleration sampling resolution, obtain the vt curve over a period of time. Integrate the vt curve to obtain the position s of the main vehicle at each moment in the future period of time.

[0122] Step S54: By integrating the vt curve, multiple behavioral trajectories of the main vehicle are obtained (i.e., the second planned trajectory in the aforementioned embodiment). These include the vehicle's velocity v, acceleration a, and position s at each moment (e.g., the 1st second, the 2nd second, ...) within a certain time period (e.g., 8 seconds).

[0123] according to Figure 5B As shown, the possible speed planning trajectory of the target vehicle or the host vehicle in the future is given by sampling and other methods. That is, by predicting the speed and acceleration, multiple speed-time curves of the target vehicle or the host vehicle are given. By integrating the different speed-time curves, multiple speed planning trajectories are obtained, which is equivalent to the second planning trajectory in the aforementioned embodiment.

[0124] In one example of this disclosure, the first planned trajectory of the obstacle is as follows: Figure 6 As shown. At times t1, t2, and t3, three first predicted trajectories are generated for the obstacle obs. The direction of the first predicted trajectory can be consistent with the velocity direction of the obstacle obs, and the length of the first predicted trajectory can be the product of the real-time velocity of the obstacle obs and the predicted time, as shown. Figure 6 The three dashed lines corresponding to t1, t2, and t3 shown in the diagram.

[0125] Reference Figure 6 At time t1, the first predicted trajectory of the obstacle obs is relatively long, which means that the predicted speed of the obstacle obs is relatively fast. There is a risk of the obstacle obs prediction line intersecting with the main vehicle ADC. After the main vehicle ADC makes a yielding decision, it records the key information and decision information of the obstacle obs in this frame.

[0126] Still refer to Figure 6 If at time t2, the predicted line of the obstacle obs becomes shorter, the predicted obstacle obs has a deceleration trend, and it is unclear whether the obstacle intends to overtake, but still retains the trend of entering the intersection area. By retaining 8 frames of the historical prediction line of the intersection frame, the main vehicle can maintain the decision to decelerate and give way.

[0127] exist Figure 6In the example shown, the obstacle's intent is further checked based on the time it takes for the current predicted line to pass the intersection point. If an obstacle OBS triggers a yield strategy for five consecutive frames (i.e., triggers an avoidance decision within five consecutive planning moments), a stop sign is established before the intersection area. Simultaneously, it continuously checks whether the obstacle OBS has left the intersection scenario. Once the obstacle OBS has moved away, the stop sign is cleared. After ensuring the main vehicle ADC is safe, it can proceed normally.

[0128] In this example, when the main vehicle (target vehicle) encounters a vehicle that needs to yield during the game with the obstacle vehicle, it can make stable yielding decisions and behaviors, and will not hesitate to yield due to changes in the predicted intent, thus causing a collision risk. When the main vehicle encounters a vehicle that meets the yielding conditions during the game with the obstacle, it will not brake suddenly due to frame-to-frame changes in the predicted line length, thus optimizing the passenger experience.

[0129] This disclosure also provides a collision avoidance device for autonomous vehicles, such as... Figure 7 As shown, it includes:

[0130] The avoidance decision module 71 is used to generate an avoidance decision indicating that the target vehicle should avoid the obstacle based on the original control trajectory of the target vehicle and the first predicted trajectory of the obstacle.

[0131] The first planned trajectory module 72 is used to generate at least one first planned trajectory for the target vehicle based on the current driving information of the target vehicle; the first planned trajectory is a trajectory that conforms to the preset speed change rules and conflicts with the avoidance decision.

[0132] Failure Operation Execution Module 73: Used to execute a failure operation on the first planned trajectory, so as to refuse to control the target vehicle according to the control operation corresponding to the first planned trajectory within the obstacle avoidance range of the target vehicle.

[0133] In one implementation, the first predicted trajectory includes multiple components, such as... Figure 8 As shown, the obstacle avoidance device for autonomous vehicles also includes:

[0134] The obstacle speed information module 81 is used to acquire obstacle speed information corresponding to each planning time point in a first set number of consecutive planning time points;

[0135] The first predicted trajectory generation module 82 is used to generate the first predicted trajectory corresponding to each planning time based on the obstacle speed information corresponding to each planning time.

[0136] In one implementation, such as Figure 9 As shown, the obstacle avoidance decision module includes:

[0137] Decision unit 91 is used to generate avoidance decisions when there is a risk of collision between the first set number of first predicted trajectories and the original control trajectory.

[0138] In one implementation, such as Figure 10 As shown, the obstacle avoidance device for autonomous vehicles also includes:

[0139] The deceleration time period module 101 is used to determine the deceleration time period based on the planning time corresponding to the second set number of first predicted trajectories when there is a risk of collision between the second set number of consecutive first predicted trajectories and the original control trajectory.

[0140] The deceleration control module 102 is used to generate control commands to control the target vehicle to perform deceleration operations during the deceleration period.

[0141] In one implementation, the current driving information includes current speed, current acceleration, and current position; such as... Figure 11 As shown, the first trajectory planning module includes:

[0142] The speed and acceleration unit 111 is used to generate at least one set of speed and acceleration planning values ​​for the target vehicle based on the current speed, current acceleration, and preset speed change rules; the speed and acceleration planning values ​​include the speed and acceleration corresponding to the planning time.

[0143] The second planning trajectory unit 112 is used to generate at least one second planning trajectory based on at least one set of velocity and acceleration planning values;

[0144] Selection unit 113 is used to select, from at least one second planning trajectory, the planning trajectory that conflicts with the avoidance decision, as the first planning trajectory.

[0145] In one implementation, the second planning trajectory unit is further used for:

[0146] In the velocity-time coordinate system, at least one time-velocity curve is generated based on at least one set of velocity and acceleration planning values ​​and at least one future planning moment;

[0147] Integrate at least one time velocity curve to obtain at least one planning position corresponding to at least one future planning moment;

[0148] A second planned trajectory is generated based on the target vehicle's current location and at least one planned location.

[0149] In one implementation, the avoidance decision is used to indicate that the target vehicle should avoid the obstacle by stopping, such as... Figure 12 As shown, the obstacle avoidance device for autonomous vehicles also includes:

[0150] The third planning trajectory module 121 is used to generate at least one third planning trajectory based on the first planning trajectory and the avoidance decision; the third planning trajectory is the trajectory in the first planning trajectory that does not conflict with the avoidance decision.

[0151] Parking area module 122 is used to determine the parking area based on the avoidance decision and the first predicted trajectory;

[0152] The parking control module 123 is used to set parking control commands in the third planned trajectory so that the target vehicle stops before reaching the parking area.

[0153] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0154] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0155] Figure 13 A schematic block diagram of an example electronic device 130 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0156] like Figure 13 As shown, device 130 includes a computing unit 131, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 132 or a computer program loaded from storage unit 138 into random access memory (RAM) 133. The RAM 133 may also store various programs and data required for the operation of device 130. The computing unit 131, ROM 132, and RAM 133 are interconnected via bus 134. Input / output (I / O) interface 135 is also connected to bus 134.

[0157] Multiple components in device 130 are connected to I / O interface 135, including: input unit 136, such as keyboard, mouse, etc.; output unit 137, such as various types of monitors, speakers, etc.; storage unit 138, such as disk, optical disk, etc.; and communication unit 139, such as network card, modem, wireless transceiver, etc. Communication unit 139 allows device 130 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0158] The computing unit 131 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 131 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 131 performs the various methods and processes described above, such as obstacle avoidance methods for autonomous vehicles. For example, in some embodiments, the obstacle avoidance methods for autonomous vehicles can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 138. In some embodiments, part or all of the computer program can be loaded and / or installed on device 130 via ROM 132 and / or communication unit 139. When the computer program is loaded into RAM 133 and executed by the computing unit 131, one or more steps of the obstacle avoidance methods for autonomous vehicles described above can be performed. Alternatively, in other embodiments, the computing unit 131 can be configured to perform obstacle avoidance methods for autonomous vehicles by any other suitable means (e.g., by means of firmware).

[0159] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0160] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0161] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0162] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0163] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0164] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0165] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0166] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for evading an obstacle by an autonomous vehicle, comprising: generating an evasion decision for the target vehicle to evade the obstacle according to a first predicted trajectory of the obstacle and an original control trajectory of the target vehicle; the first predicted trajectory comprises a plurality of; generating at least one first planned trajectory of the target vehicle based on current driving information of the target vehicle; the first planned trajectory is a trajectory that complies with a preset speed variation rule and conflicts with the evasion decision; performing a failure operation on the first planned trajectory to reject controlling the target vehicle according to a control operation corresponding to the first planned trajectory within an evasion range of the target vehicle to the obstacle; the failure operation at least includes a temporary deletion operation or a marking operation; the evasion range includes a distance range and a time range; the method further comprises: obtaining obstacle speed information corresponding to each planning time in a continuous first set number of planning times; generating the first predicted trajectory corresponding to each planning time according to the obstacle speed information corresponding to each planning time.

2. The method of claim 1, wherein, the generating an evasion decision for the target vehicle to evade the obstacle according to a first predicted trajectory of the obstacle and an original control trajectory of the target vehicle, comprising: generating the evasion decision in the case that the first set number of first predicted trajectories and the original control trajectory have a collision risk.

3. The method of claim 1, further comprising: determining a deceleration time period based on the planning time corresponding to the second set number of first predicted trajectories in the case that the continuous second set number of first predicted trajectories and the original control trajectory have a collision risk; generating a control instruction for controlling the target vehicle to perform a deceleration operation within the deceleration time period.

4. The method of any of claims 1-3, wherein, the current driving information includes current speed, current acceleration, and current position; the generating at least one first planned trajectory of the target vehicle based on current driving information of the target vehicle, comprising: generating at least one set of speed and acceleration planning values of the target vehicle according to the current speed, the current acceleration, and the preset speed variation rule; the speed and acceleration planning values include speed and acceleration corresponding to a planning time; generating at least one second planned trajectory according to the at least one set of speed and acceleration planning values; selecting a planning trajectory that conflicts with the evasion decision from the at least one second planned trajectory as the first planned trajectory.

5. The method of claim 4, wherein, the generating at least one second planned trajectory according to the at least one set of speed and acceleration planning values, comprising: generating at least one time-speed curve in a speed-time coordinate system according to the at least one set of speed and acceleration planning values and at least one planning time; integrating the at least one time-speed curve to obtain at least one planning position corresponding to at least one planning time; generating the second planned trajectory according to the current position of the target vehicle and the at least one planning position.

6. The method of claim 5, wherein, the evasion decision is used to indicate that the target vehicle evades by parking, and the method further comprises: generating at least one third planning trajectory according to the first planning trajectory and the avoidance decision; the third planning trajectory is a trajectory in the first planning trajectory that does not conflict with the avoidance decision; determining a parking area according to the avoidance decision and the first predicted trajectory; setting a parking control instruction in the third planning trajectory, so that the target vehicle stops before reaching the parking area.

7. An avoidance device of an autonomous vehicle, comprising: an avoidance decision module configured to generate an avoidance decision indicating that the target vehicle avoids an obstacle according to an original control trajectory of the target vehicle and a first predicted trajectory of the obstacle; the first predicted trajectory comprises a plurality of; a first planning trajectory module configured to generate at least one first planning trajectory of the target vehicle based on current driving information of the target vehicle; the first planning trajectory is a trajectory that meets a preset speed change rule and conflicts with the avoidance decision; a failure operation execution module configured to perform a failure operation on the first planning trajectory to refuse to control the target vehicle according to a control operation corresponding to the first planning trajectory within an avoidance range of the target vehicle to the obstacle; the failure operation at least includes a temporary deletion operation, a direct deletion operation or a marking operation; the avoidance range includes a distance range and a time range; the device further comprises: an obstacle speed information module configured to obtain obstacle speed information corresponding to each planning moment in a continuous first set number of planning moments; a first predicted trajectory generation module configured to generate the first predicted trajectory corresponding to each planning moment according to the obstacle speed information corresponding to each planning moment.

8. The apparatus of claim 7, wherein, the avoidance decision module comprises: a decision unit configured to generate the avoidance decision in a case where the first set number of first predicted trajectories and the original control trajectory have a collision risk.

9. The device of claim 7, further comprising: a deceleration time period module configured to determine a deceleration time period based on a planning moment corresponding to the second set number of first predicted trajectories in a case where the continuous second set number of first predicted trajectories and the original control trajectory have a collision risk; a deceleration control module configured to generate a control instruction for controlling the target vehicle to perform a deceleration operation within the deceleration time period.

10. The apparatus of any of claims 7-9, wherein, the current driving information includes a current speed, a current acceleration and a current position; the first planning trajectory module comprises: a speed and acceleration unit configured to generate at least one set of speed and acceleration planning values of the target vehicle according to the current speed, the current acceleration and a preset speed change rule; the speed and acceleration planning values include speed and acceleration corresponding to a planning moment; a second planning trajectory unit configured to generate at least one second planning trajectory according to the at least one set of speed and acceleration planning values; a selection unit configured to select a planning trajectory that conflicts with the avoidance decision from the at least one second planning trajectory as the first planning trajectory.

11. The apparatus of claim 10, wherein, the second planning trajectory unit is further configured to: In a speed-time coordinate system, at least one time-speed curve is generated according to the at least one set of speed and acceleration planning values and the at least one planning time; At least one planning position corresponding to the at least one planning time is obtained by integrating the at least one time-speed curve; The second planning trajectory is generated according to the current position of the target vehicle and the at least one planning position.

12. The apparatus of claim 11, wherein, The avoidance decision is used to represent that the target vehicle is avoided by parking, and the device further comprises: A third planning trajectory module is configured to generate at least one third planning trajectory according to the first planning trajectory and the avoidance decision; the third planning trajectory is a trajectory in the first planning trajectory that does not conflict with the avoidance decision; A parking area module is configured to determine a parking area according to the avoidance decision and the first prediction trajectory; A parking control module is configured to set a parking control instruction in the third planning trajectory, so that the target vehicle stops before reaching the parking area.

13. An electronic device comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

14. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to make the computer execute the method according to any one of claims 1-6.

15. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions, when executed by the processor, implement the steps of the method of any one of claims 1-6.

Images