CN111062372B

CN111062372B - Method and device for predicting obstacle track

Info

Publication number: CN111062372B
Application number: CN202010172799.8A
Authority: CN
Inventors: 任冬淳; 夏华夏; 樊明宇; 钱德恒; 丁曙光
Original assignee: Beijing Sankuai Online Technology Co Ltd
Current assignee: Beijing Sankuai Online Technology Co Ltd
Priority date: 2020-03-13
Filing date: 2020-03-13
Publication date: 2020-07-03
Anticipated expiration: 2040-03-13
Also published as: CN111062372A

Abstract

The specification discloses a method and a device for predicting barrier tracks, which can determine track errors between historical real tracks and historical predicted tracks according to the acquired historical real tracks of a barrier and a plurality of historical predicted tracks of the barrier output by a track planning algorithm, wherein the track errors are smaller when the quantity of the historical predicted tracks is larger, the total area occupied by the historical predicted tracks is determined, and the total area is larger when the quantity of the historical predicted tracks is larger. Thus, a predicted trajectory score may be determined based on the trajectory error and the total area, where the predicted trajectory score is negatively correlated with the trajectory error and the predicted trajectory score is negatively correlated with the total area. The number of historical predicted trajectories output by the trajectory prediction algorithm when the predicted trajectory score is maximum can be determined, and the trajectory of the obstacle can be predicted according to the determined number. The method solves the balance problem of the influence of the precision of the predicted track of the obstacle and the predicted track of the obstacle on the self track planned by the unmanned equipment in the prior art.

Description

Method and device for predicting obstacle track

Technical Field

The application relates to the technical field of unmanned driving, in particular to a method and a device for predicting an obstacle trajectory.

Background

With the intensive research in the field of unmanned driving, trajectory planning of unmanned equipment becomes more and more important, and when the trajectory of the unmanned equipment is planned, obstacles in the surrounding environment of the unmanned equipment need to be avoided so as to avoid accidents.

When the unmanned equipment performs trajectory planning in a moving state, obstacles (mainly including dynamic obstacles such as vehicles and pedestrians) also move continuously. Therefore, the unmanned aerial vehicle can predict the predicted trajectory of the obstacle according to the historical trajectory of the obstacle and the surrounding environment, and plan the trajectory of the unmanned aerial vehicle by avoiding the predicted trajectory of the obstacle. The drone may use different obstacle trajectory prediction methods to predict a predicted trajectory of an obstacle. For example, the historical trajectory of the obstacle may be input into a long-short term memory (LSTM) model to obtain a plurality of predicted trajectories of the obstacle output by the LSTM model, or, for example, the historical trajectory of the obstacle may be described as a trajectory image, and the trajectory image may be input into a Convolutional Neural Network (CNN) to obtain a plurality of predicted trajectories of the obstacle output by the CNN model.

If the predicted track number of the obstacle predicted by the unmanned equipment is more, the predicted track is more likely to be overlapped with the real track of the future movement of the obstacle, namely, the higher the precision of the track of the obstacle predicted by the unmanned equipment is, the better the unmanned equipment plans the track of the unmanned equipment. However, the larger the number of predicted trajectories for which the unmanned aerial vehicle predicts the obstacle, the smaller the area in which the unmanned aerial vehicle can plan its own trajectory, and the larger the influence on the unmanned aerial vehicle in planning its own trajectory.

Therefore, how to balance the accuracy of the predicted trajectory of the obstacle and the influence of the predicted trajectory of the obstacle on the unmanned equipment planning trajectory is a problem to be solved urgently.

Disclosure of Invention

The embodiments of the present disclosure provide a method and an apparatus for predicting an obstacle trajectory, so as to partially solve the above problems in the prior art.

The embodiment of the specification adopts the following technical scheme:

the present specification provides a method for predicting an obstacle trajectory, the method comprising:

obtaining historical real tracks of the obstacles and historical results of primary prediction of the obstacles through a track prediction algorithm, wherein the results comprise a plurality of historical predicted tracks output by the track prediction algorithm;

determining track errors of the historical real tracks and the historical predicted tracks and total area occupied by the historical predicted tracks, wherein the track errors are in negative correlation with the quantity of the historical predicted tracks output by the track prediction algorithm, and the total area is in positive correlation with the quantity of the historical predicted tracks output by the track prediction algorithm;

determining a predicted trajectory score based on the determined trajectory errors and the total area, wherein the predicted trajectory score is negatively correlated with the trajectory errors and the predicted trajectory score is negatively correlated with the total area;

determining the number of historical predicted tracks output by the track prediction algorithm when the predicted track score is maximum, and taking the number as the predicted number;

predicting a trajectory of the obstacle based on the determined number of predictions.

Optionally, determining a trajectory error between the historical real trajectory and each historical predicted trajectory specifically includes:

determining the error between the historical real track and the historical predicted track as a reference error for each historical predicted track;

and taking the minimum value of the reference errors determined for each historical predicted track as the track error between the historical real track and each historical predicted track.

Optionally, the method specifically includes:

determining the projection area of the obstacle on the ground and the position information of each historical predicted track;

and determining the total area occupied by each historical predicted track according to the projection area and the position information of each historical predicted track.

Optionally, determining a total area occupied by each historical predicted track according to the projection area and the position information of each historical predicted track, specifically including:

determining a position union set of all historical predicted tracks according to the position information of each historical predicted track;

and determining the total area occupied by each historical predicted track according to the projection area and the position union set.

Optionally, obtaining a result of predicting the obstacle once through a trajectory prediction algorithm historically includes:

acquiring the historical state of the obstacle;

and inputting the historical states into the track prediction algorithm to obtain a plurality of historical predicted tracks output by predicting the obstacle once through the track prediction algorithm.

Optionally, determining the number of historical predicted trajectories output by the trajectory prediction algorithm when the predicted trajectory score is maximum, as the predicted number, specifically includes:

adjusting parameters of the trajectory prediction algorithm to change the number of historical predicted trajectories output by the trajectory prediction algorithm;

taking the number of historical predicted tracks output by the track prediction algorithm after the parameters are adjusted as an undetermined number;

for each undetermined quantity, determining the grade of the predicted track according to the undetermined quantity of historical predicted tracks output by the track prediction algorithm;

and taking the number to be determined when the predicted track score is maximum as the predicted number.

Optionally, predicting the trajectory of the obstacle according to the determined prediction number specifically includes:

acquiring the state of the obstacle;

and inputting the states of the obstacles into the trajectory prediction algorithm according to the determined prediction number to obtain the trajectories of the prediction number output by the trajectory prediction algorithm.

The present specification provides an apparatus for predicting an obstacle trajectory, the apparatus comprising:

the acquisition module is used for acquiring historical real tracks of the obstacles and results of one-time prediction of the obstacles in history through a track prediction algorithm, wherein the results comprise a plurality of historical predicted tracks output by the track prediction algorithm;

the first determination module is used for determining track errors of the historical real tracks and the historical predicted tracks and the total area occupied by the historical predicted tracks, wherein the track errors are in negative correlation with the quantity of the historical predicted tracks output by the track prediction algorithm, and the total area is in positive correlation with the quantity of the historical predicted tracks output by the track prediction algorithm;

a second determining module, configured to determine a predicted trajectory score according to the determined trajectory error and the total area, where the predicted trajectory score is negatively correlated with the trajectory error and the predicted trajectory score is negatively correlated with the total area;

a third determining module, configured to determine, as a predicted number, the number of historical predicted tracks output by the track prediction algorithm when the predicted track score is maximum;

a prediction module to predict a trajectory of the obstacle based on the determined number of predictions.

The present specification provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the above-described method of predicting an obstacle trajectory.

The present specification provides an unmanned aerial vehicle, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the above method for predicting an obstacle trajectory when executing the program.

The embodiment of the specification adopts at least one technical scheme which can achieve the following beneficial effects:

the method comprises the steps of obtaining historical real tracks of the obstacles and a plurality of historical predicted tracks of the obstacles output by a track planning algorithm, determining track errors of the historical real tracks and the historical predicted tracks, wherein the track errors are smaller when the quantity of the historical predicted tracks is larger, the total area occupied by the historical predicted tracks is determined, and the total area is larger when the quantity of the historical predicted tracks is larger. Therefore, a predicted trajectory score may be determined based on the trajectory errors and the total area, wherein the predicted trajectory score is negatively correlated with the trajectory errors and the predicted trajectory score is negatively correlated with the total area, and the number of historical predicted trajectories output by the trajectory prediction algorithm when the predicted trajectory score is maximum is determined as the predicted number. Finally, the trajectory of the obstacle may be predicted based on the determined number of predictions. The prediction number determined by the method not only considers the error between the historical predicted track of the obstacle and the historical real track, but also considers the total area occupied by the historical predicted track of the obstacle, thereby solving the balance problem of the influence of the precision of the predicted track of the obstacle and the predicted track of the obstacle on the self track planned by the unmanned equipment in the prior art, and realizing the beneficial effect of adaptively adjusting the number of the predicted tracks of the obstacle to balance the precision of the predicted track of the obstacle and the influence of the predicted track of the obstacle on the self track planned by the unmanned equipment.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

fig. 1 is a flowchart of a method for predicting an obstacle trajectory according to an embodiment of the present disclosure;

FIG. 2 is a schematic diagram of a plurality of historical predicted tracks provided by an embodiment of the present description;

FIG. 3 is a diagram illustrating a union of locations of all historical predicted tracks in an embodiment of the present disclosure;

fig. 4 is a schematic structural diagram of an apparatus for predicting an obstacle trajectory according to an embodiment of the present disclosure;

fig. 5 is a schematic diagram of an unmanned aerial vehicle corresponding to fig. 1 provided in an embodiment of the present specification.

Detailed Description

In order to make the objects, technical solutions and advantages of the present disclosure more apparent, the technical solutions of the present disclosure will be clearly and completely described below with reference to the specific embodiments of the present disclosure and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments in the present specification without any inventive step are within the scope of the present application.

The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

Fig. 1 is a flowchart of a method for predicting an obstacle trajectory according to an embodiment of the present disclosure, which may specifically include the following steps:

s100: obtaining historical real tracks of the obstacles and historical results of primary prediction of the obstacles through a track prediction algorithm, wherein the results comprise a plurality of historical predicted tracks output by the track prediction algorithm.

The method for predicting the trajectory of the obstacle provided by the specification can be applied to unmanned driving, and particularly, when the trajectory of the unmanned equipment is planned, the unmanned equipment needs to avoid the obstacle in the surrounding environment, needs to predict the trajectory of the obstacle, and plans the trajectory of the unmanned equipment according to the predicted trajectory of the obstacle. When the unmanned equipment predicts the track of the obstacle, the number of predicted tracks of the obstacle needs to be determined, and meanwhile the accuracy of the predicted obstacle track and the influence of the predicted obstacle track on the unmanned equipment planning self track are considered. The unmanned device may determine a number of predicted trajectories of the obstacle based on the historical true trajectory and the historical predicted trajectory of the obstacle.

First, the unmanned device may acquire a historical true trajectory of the obstacle.

Specifically, the unmanned device can acquire data of actual movement of the obstacle, and determine a historical real track of the obstacle according to the data of the actual movement of the obstacle. In addition, the unmanned device may determine the historical actual trajectory of the obstacle by using a map difference method, a point cloud clustering method, and the like according to data acquired by sensors such as an image sensor and a laser radar, which is not described herein again.

Then, the unmanned device can obtain the result of predicting the obstacle once through the trajectory prediction algorithm historically, and the result comprises a plurality of historical predicted trajectories output by the trajectory prediction algorithm.

Specifically, the unmanned device may obtain a plurality of historical predicted trajectories that have been historically predicted for the obstacle by trajectory prediction algorithms such as LSTM models, CNN models, and the like. It should be noted here that predicted trajectories output by different trajectory prediction algorithms are not exactly the same, and the optimal predicted number of predicted trajectories of the obstacle corresponding to different trajectory prediction algorithms may not necessarily be the same, so that the acquired historical predicted trajectories of the plurality of obstacles are all output by the same trajectory prediction algorithm at the same time.

In addition, the unmanned equipment can also acquire the historical state of the obstacle, input the historical state into a track prediction algorithm, and obtain a plurality of historical prediction tracks output by predicting the obstacle once through the track prediction algorithm. That is, the unmanned device may not directly obtain the historical predicted trajectory, but obtain the historical state of the obstacle, input the obtained historical state of the obstacle and historical ambient environment information into the trajectory prediction algorithm, and obtain a result of the trajectory prediction algorithm predicting the obstacle once, where the predicted result includes a plurality of historical predicted trajectories. The unmanned equipment can determine the historical state of the obstacle and historical surrounding environment information according to data collected by sensors such as an image sensor and a laser radar.

Of course, the unmanned device may also obtain historical real trajectories of multiple obstacles, as well as several historical predicted trajectories for each obstacle. When the unmanned equipment acquires a plurality of historical predicted tracks of each obstacle, the unmanned equipment outputs a plurality of historical predicted results of the obstacle by using the same track prediction algorithm for each obstacle.

S102: and determining track errors of the historical real tracks and the historical predicted tracks and the total area occupied by the historical predicted tracks, wherein the track errors are in negative correlation with the quantity of the historical predicted tracks output by the track prediction algorithm, and the total area is in positive correlation with the quantity of the historical predicted tracks output by the track prediction algorithm.

From the historical real trajectory of the obstacle and the plurality of historical predicted trajectories acquired in step S100, trajectory errors between the historical real trajectory and each historical predicted trajectory may be determined, where the trajectory errors are inversely related to the number of historical predicted trajectories output by the trajectory prediction algorithm.

First, for each historical predicted trajectory, an error of the historical real trajectory from the historical predicted trajectory may be determined as a reference error.

Specifically, for each historical predicted track, the same number of sampling points are respectively determined from the historical real track and the historical predicted track. And determining the average displacement error between the historical real track and the historical predicted track according to the determined position information of each sampling point, and taking the average displacement error as a reference error. If it is in the history of real track

A sampling point is represented as

The first in the historical predicted trajectory

A sampling point is represented as

Then the reference error between the historical real track and the historical predicted track

As shown in equation (1).

（1）

Wherein the content of the first and second substances,

and the number of sampling points in the historical real track or the historical prediction track is used as the number of the sampling points.

Then, the minimum value of the reference errors determined for each of the historical predicted trajectories may be used as the trajectory error of the historical real trajectory from the respective historical predicted trajectory.

Specifically, if the number of the historical predicted tracks is

In respect of

A reference error determined from the historical predicted trajectory is

Then the track error between the historical real track and each historical predicted track

As shown in equation (2).

（2）

It should be noted here that the number of tracks is predicted when history is used

When the historical predicted track is larger, the historical predicted track is more likely to be overlapped with the historical real track, and the track error between the historical real track and each historical predicted track

The closer to 0. That is, the number of historical predicted tracks

The larger the error between the historical true trajectory and each historical predicted trajectory

The smaller, the track error

Number of historical predicted tracks output from track prediction algorithm

A negative correlation.

While determining the track error of the historical real track and each historical predicted track, the unmanned device can also determine the total area occupied by each historical predicted track. Wherein the total area is positively correlated with the number of historical predicted trajectories output by the trajectory prediction algorithm.

First, the projected area of the obstacle on the ground can be determined, as well as the position information for each historical predicted trajectory. And secondly, determining the total area occupied by each historical predicted track according to the projection area and the position information of each historical predicted track.

Specifically, when the unmanned device determines the projection area of the obstacle on the ground, the unmanned device may use the area in the bounding box as the projection area of the obstacle on the ground according to information of a bounding box (bounding box) of the obstacle output by a sensing module of the unmanned device, where the bounding box is a minimum external region of the obstacle and the surrounding environment. The method can also process the acquired data according to the data acquired by sensors such as an image sensor, a laser radar and the like, determine the information of the minimum external region of the obstacle and the surrounding environment, and take the area of the minimum external region as the projection area of the obstacle on the ground. The drone may also determine location information for each historical predicted trajectory while determining the projected area of the obstacle on the ground. If the history predicted track is historical

Is timed to

The predicted track of the movement of the obstacle between the moments, the position information of the historical predicted track can be that the obstacle is at

Is timed to

The position between the moments and the unmanned equipment is

Relative positional relationship of the positions of the time of day. On the map, the historical predicted track can be represented as a continuous area with width information, and the position information of the historical predicted track can also be represented as map coordinate information.

Fig. 2 is a schematic diagram of a plurality of historical predicted tracks provided by an embodiment of the present disclosure. In fig. 2, a is an unmanned device, B is an obstacle, B1, B2, B3 are three historical predicted trajectories of the obstacle B, respectively, and B1, B2, B3 are represented by different frame lines, respectively, so that the position information of the historical predicted trajectories of the obstacle can be as shown in fig. 2.

After the projection area and the position information of each historical prediction track are determined, the position union of all historical prediction tracks can be determined according to the position information of each historical prediction track, and the total area occupied by each historical prediction track can be determined according to the projection area and the position union.

Since each historical predicted track may have a partial overlapping area on the map, the position union of all the historical predicted tracks can be determined according to the information of each historical predicted track. Fig. 3 is a schematic diagram of a position union set of all historical predicted tracks in the embodiment of the present specification, and following the above example, in fig. 2, the historical predicted tracks B1, B2, B3 have partial overlapping areas, and then the position union set of all historical predicted tracks of the obstacle B may be as shown by the shaded area in fig. 3. The total area occupied by each historical predicted track

As shown in equation (3).

（3）

Wherein the content of the first and second substances,

representing the union of the areas occupied by all historical predicted traces,

is shown in

The area of the obstacle at the moment of time,

indicating an obstacle at the first

In a historical predicted track

The smallest circumscribed area of the moment.

In addition, the total area occupied by each historical predicted track other than that determined by equation (3)

Besides, a union of projection areas of the obstacles in the historical predicted tracks at the same time can be determined, and the total area occupied by the historical predicted tracks can be determined according to the determined union of the projection areas, which can be shown in formula (4).

（4）

In the above formula (3) and formula (4), the obstacle is in the second place

In a historical predicted track

Minimum circumscribed area of time

The information of the bounding box determined by the perception module of the unmanned device may be changed or may be a fixed value, for example, the average value of the minimum circumscribed area of the obstacle in a historical period of time may be used as the average value

。

S104: determining a predicted trajectory score based on the determined trajectory errors and the total area, wherein the predicted trajectory score is negatively correlated with the trajectory errors and the predicted trajectory score is negatively correlated with the total area.

Determination of the trajectory error of the obstacle by the above-described step S102

And total area occupied by each historical predicted track

The drone may then determine a predicted trajectory score for the obstacle

Predictive trajectory scoring of an obstacle

As shown in equation (5).

（5）

In equation (5), the number of trajectories is predicted when the history of the obstacle is

The larger the error in the trajectory of the obstacle

Smaller, smaller molecules

The larger the predicted trajectory score

The larger, i.e. predicted trajectory score of the obstacle

Error of trajectory with obstacle

A negative correlation. Number of historical predicted trajectories when obstacles

The larger the sizeTotal area of each history prediction track of denominator

The larger the predicted trajectory score of the obstacle

The total area occupied by each historical predicted track

A negative correlation.

S106: and determining the number of historical predicted tracks output by the track prediction algorithm when the predicted track score is maximum as a predicted number.

The smaller the trajectory error of the obstacle, the greater the accuracy of the predicted trajectory of the obstacle. The larger the total area occupied by each historical predicted track of the obstacle is, the larger the influence of the predicted track of the obstacle on the unmanned equipment planning self track is, so that the formula (5) determines the predicted track score of the obstacle, not only the accuracy of the predicted track of the obstacle is considered, but also the influence of the predicted track of the obstacle on the unmanned equipment planning self track is considered, and when the predicted track score of the obstacle in the formula (5) takes the maximum value, the beneficial effect that the influence on the unmanned equipment planning self track is smaller under the condition that the accuracy of the predicted track of the obstacle is higher can be achieved.

Because the result of the primary prediction of the obstacle by the trajectory prediction algorithm may include a plurality of historical predicted trajectories, in order to determine the number (i.e., the predicted number) of the historical predicted trajectories output by the trajectory prediction algorithm when the predicted trajectory score is the maximum, the parameters of the trajectory prediction algorithm may be adjusted to change the number of the historical predicted trajectories output by the trajectory prediction algorithm, and after the parameters are adjusted, the number of the historical predicted trajectories output by the trajectory prediction algorithm is used as the undetermined number, and for each undetermined number, the predicted trajectory score may be determined according to the undetermined number of historical predicted trajectories output by the trajectory prediction algorithm, and the undetermined number when the predicted trajectory score is the maximum is used as the predicted number.

Specifically, after the unmanned equipment adjusts the parameter value of the trajectory prediction algorithm each time, the trajectory prediction algorithm can output different quantities of historical predicted trajectories. For each parameter adjustment, the undetermined number of historical predicted tracks corresponding to the parameter adjustment can be determined, the undetermined number of historical predicted tracks corresponding to the parameter adjustment can be adjusted according to the determined parameter, and the score of the adjusted predicted track can be determined according to a formula (5)

. Obtained by adjusting parameters several times

In (1), can determine

Corresponding to

As the predicted number.

In addition, in determining the number of predictions

When the value is taken, historical real tracks of a plurality of obstacles and a plurality of historical predicted tracks corresponding to each obstacle can be obtained, and the sum of the predicted track scores of each obstacle can be determined to be used as the final predicted track score of all the obstacles. The final predicted trajectory score for all obstacles

As shown in equation (6).

（6）

Wherein the content of the first and second substances,

as to the number of the obstacles,

is as follows

The predicted trajectory score for each obstacle.

At a certain moment, the number of obstacles in the environment surrounding the unmanned aerial device is fixed, that is,

may be constant. Aiming at each obstacle, the undetermined number of historical predicted trajectories of the obstacle can be obtained by adjusting parameters of a trajectory prediction algorithm, and then the final predicted trajectory scores of all the obstacles obtained by adjusting for a plurality of times can be determined

Corresponding to

As the predicted number.

S108: predicting a trajectory of the obstacle based on the determined number of predictions.

After the predicted number is determined in step S106, the unmanned device may obtain the state of the obstacle, and then, according to the determined predicted number, input the state of the obstacle and the historical real trajectory of the obstacle into the trajectory prediction algorithm to obtain the trajectory of the predicted number output by the trajectory prediction algorithm. In addition, when the drone detects the presence of multiple obstacles in the surrounding environment, a predicted number of trajectories for each obstacle may be determined. The unmanned equipment can plan the track of the unmanned equipment according to the determined tracks of the predicted number of obstacles.

In the method for predicting the trajectory of the obstacle provided by the present specification, the historical predicted trajectories of the obstacle are all determined by the same trajectory prediction algorithm, and therefore, the obtained predicted number is the number of the trajectories of the obstacle corresponding to the trajectory prediction algorithm. Different trajectory prediction algorithms may each determine the corresponding prediction number by the above-described method. This description is not repeated.

The method for predicting the obstacle trajectory provided by the present specification is particularly applicable to the field of delivery using an unmanned device, for example, a delivery scene such as express delivery and takeout using an unmanned device. Specifically, in the above-described scenario, delivery may be performed using an unmanned vehicle fleet configured with a plurality of unmanned devices.

Based on the method for predicting the obstacle trajectory shown in fig. 1, the embodiment of the present specification further provides a schematic structural diagram of an apparatus for predicting the obstacle trajectory, as shown in fig. 4.

Fig. 4 is a schematic structural diagram of an apparatus for predicting an obstacle trajectory according to an embodiment of the present disclosure, where the apparatus includes:

an obtaining module 401, configured to obtain a historical real trajectory of an obstacle, and a result of predicting the obstacle once through a trajectory prediction algorithm in history, where the result includes a plurality of historical predicted trajectories output by the trajectory prediction algorithm;

a first determining module 402, configured to determine track errors of the historical real tracks and the historical predicted tracks, and a total area occupied by the historical predicted tracks, where the track errors are negatively related to the number of the historical predicted tracks output by the track prediction algorithm, and the total area is positively related to the number of the historical predicted tracks output by the track prediction algorithm;

a second determining module 403, configured to determine a predicted trajectory score according to the determined trajectory error and the total area, where the predicted trajectory score is negatively related to the trajectory error, and the predicted trajectory score is negatively related to the total area;

a third determining module 404, configured to determine, as a predicted number, the number of historical predicted tracks output by the track prediction algorithm when the predicted track score is the maximum;

a prediction module 405 for predicting a trajectory of the obstacle based on the determined number of predictions.

Optionally, the first determining module 402 is specifically configured to, for each historical predicted track, determine an error between the historical real track and the historical predicted track as a reference error; and taking the minimum value of the reference errors determined for each historical predicted track as the track error between the historical real track and each historical predicted track.

Optionally, the first determining module 402 is specifically configured to determine a projected area of the obstacle on the ground and position information of each historical predicted trajectory; and determining the total area occupied by each historical predicted track according to the projection area and the position information of each historical predicted track.

Optionally, the first determining module 402 is specifically configured to determine a position union set of all historical predicted trajectories according to the position information of each historical predicted trajectory; and determining the total area occupied by each historical predicted track according to the projection area and the position union set.

Optionally, the obtaining module 401 is specifically configured to obtain a historical state of the obstacle; and inputting the historical states into the track prediction algorithm to obtain a plurality of historical predicted tracks output by predicting the obstacle once through the track prediction algorithm.

Optionally, the third determining module 404 is specifically configured to adjust a parameter of the trajectory prediction algorithm to change the number of historical predicted trajectories output by the trajectory prediction algorithm; taking the number of historical predicted tracks output by the track prediction algorithm after the parameters are adjusted as an undetermined number; for each undetermined quantity, determining the grade of the predicted track according to the undetermined quantity of historical predicted tracks output by the track prediction algorithm; and taking the number to be determined when the predicted track score is maximum as the predicted number.

Optionally, the prediction module 405 is specifically configured to obtain a state of the obstacle; and inputting the states of the obstacles into the trajectory prediction algorithm according to the determined prediction number to obtain the trajectories of the prediction number output by the trajectory prediction algorithm.

Embodiments of the present specification also provide a computer-readable storage medium, which stores a computer program, and the computer program can be used to execute the method for predicting the obstacle trajectory provided in fig. 1.

Based on the method for predicting the obstacle trajectory shown in fig. 1, the embodiment of the present specification also proposes a schematic structural diagram of the unmanned device shown in fig. 5. As shown in fig. 5, the drone includes, at the hardware level, a processor, an internal bus, a network interface, a memory, and a non-volatile memory, although it may also include hardware required for other services. The processor reads a corresponding computer program from the non-volatile memory into the memory and then runs the computer program to implement the method for predicting the obstacle trajectory described above with reference to fig. 1.

Of course, besides the software implementation, the present specification does not exclude other implementations, such as logic devices or a combination of software and hardware, and the like, that is, the execution subject of the following processing flow is not limited to each logic unit, and may be hardware or logic devices.

In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Language Description Language), traffic, pl (core unified Programming Language), HDCal, JHDL (Java Hardware Description Language), langue, Lola, HDL, laspam, hardsradware (Hardware Description Language), vhjhd (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functions of the various elements may be implemented in the same one or more software and/or hardware implementations of the present description.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

As will be appreciated by one skilled in the art, embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the description may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

This description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The above description is only an example of the present specification, and is not intended to limit the present specification. Various modifications and alterations to this description will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification should be included in the scope of the claims of the present specification.

Claims

1. A method of predicting an obstacle trajectory, the method comprising:

2. The method of claim 1, wherein determining the trajectory error of the historical real trajectory and each historical predicted trajectory specifically comprises:

3. The method of claim 1, wherein determining a total area occupied by each historical predicted trajectory comprises:

4. The method of claim 3, wherein determining a total area occupied by each historical predicted trajectory based on the projected area and the location information of each historical predicted trajectory comprises:

5. The method of claim 1, wherein obtaining a result of a historical one-time prediction of the obstacle by a trajectory prediction algorithm comprises:

acquiring the historical state of the obstacle;

6. The method according to claim 5, wherein determining the number of historical predicted trajectories output by the trajectory prediction algorithm when the predicted trajectory score is the maximum as the predicted number specifically comprises:

7. The method of claim 1, wherein predicting the trajectory of the obstacle based on the determined number of predictions comprises:

acquiring the state of the obstacle;

8. An apparatus for predicting an obstacle trajectory, the apparatus comprising:

9. A computer-readable storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the method of any of the preceding claims 1-7.

10. An unmanned aerial device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the method of any of claims 1-7.