CN117152708A - Spoofing attack detection method, device, equipment and storage medium - Google Patents
Spoofing attack detection method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN117152708A CN117152708A CN202210567504.6A CN202210567504A CN117152708A CN 117152708 A CN117152708 A CN 117152708A CN 202210567504 A CN202210567504 A CN 202210567504A CN 117152708 A CN117152708 A CN 117152708A
- Authority
- CN
- China
- Prior art keywords
- detection
- laser radar
- cloud data
- point cloud
- assistance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 235
- 238000003860 storage Methods 0.000 title claims abstract description 17
- 230000002159 abnormal effect Effects 0.000 claims abstract description 110
- 230000004044 response Effects 0.000 claims abstract description 90
- 238000000034 method Methods 0.000 claims abstract description 26
- 230000015654 memory Effects 0.000 claims description 29
- 238000004590 computer program Methods 0.000 claims description 18
- 230000008859 change Effects 0.000 claims description 9
- 238000010586 diagram Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000002547 anomalous effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/4802—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Traffic Control Systems (AREA)
Abstract
The specification relates to the technical field of automatic driving, and provides a spoofing attack detection method, a spoofing attack detection device and a spoofing attack detection storage medium, wherein the method comprises the following steps: acquiring point cloud data currently acquired by a laser radar of the vehicle; identifying whether an abnormal object exists in the point cloud data; broadcasting an auxiliary detection request carrying point cloud data and a space position corresponding to the abnormal object to surrounding vehicles when the abnormal object exists in the point cloud data; receiving an assistance detection response returned by the surrounding vehicles aiming at the assistance detection request; and determining whether the laser radar has a spoofing attack according to the assistance detection response. According to the embodiment of the specification, the detection accuracy of the spoofing attack aiming at the laser radar can be improved, and the driving safety of the automatic driving vehicle is improved.
Description
Technical Field
The present disclosure relates to the field of autopilot technologies, and in particular, to a method, an apparatus, a device, and a storage medium for detecting a spoofing attack.
Background
Lidar is an indispensable driving environment sensor in the sensing module of the autopilot system, which can provide real-time three-dimensional (3D) data of the surrounding environment for the car. Meanwhile, due to the characteristic that the laser radar actively emits light beams, the camera is not easily affected by surrounding environment (such as weak light) compared with a camera, and an object detection module of the automatic driving system can also directly utilize 3D data collected by the laser radar to conduct object identification. Some students in recent years have proposed using physical devices to fool lidar in the manner of an insertion point. Because of the limitations of the object detection model, the attacks can often cause an automatic driving system of the victim vehicle to misuse the object to appear nearby by inserting a small number of points, so that actions such as sudden braking and the like are performed, and the driving safety of the automatic driving vehicle is further affected.
Disclosure of Invention
An object of the embodiments of the present disclosure is to provide a method, an apparatus, a device, and a storage medium for detecting a fraud attack, so as to improve accuracy of detecting a fraud attack against a lidar.
In order to achieve the above object, in one aspect, an embodiment of the present disclosure provides a fraud attack detection method, including:
acquiring point cloud data currently acquired by a laser radar of the vehicle;
identifying whether an abnormal object exists in the point cloud data;
broadcasting an auxiliary detection request carrying point cloud data and a space position corresponding to the abnormal object to surrounding vehicles when the abnormal object exists in the point cloud data;
receiving an assistance detection response returned by the surrounding vehicles aiming at the assistance detection request;
and determining whether the laser radar has a spoofing attack according to the assistance detection response.
In the fraud attack detection method of the embodiment of the present disclosure, the identifying whether an abnormal object exists in the point cloud data includes:
inputting the point cloud data into a region candidate network, and predicting to obtain an object candidate frame set;
determining whether the point number of the candidate object in each object candidate frame in the object candidate frame set is positioned in a point number space range;
When the number of points of the candidate object in the object candidate frame is not in the number space range, judging whether other object candidate frames exist in the distance range from the candidate object to the laser radar;
and if no other object candidate frame exists in the distance range from the object candidate to the laser radar, confirming that the object is an abnormal object.
In the fraud attack detection method of the embodiment of the present disclosure, the point space range includes a point space located between the first regression curve and the second regression curve;
the first regression curve is a regression curve of the minimum value of the point number of the object in the laser radar field of view along with the change of the detection distance under the normal condition;
and the second regression curve is a regression curve of the maximum value of the point number of the object in the laser radar field of view along with the change of the detection distance under the normal condition.
In the fraud attack detection method of the embodiment of the present disclosure, the determining whether the laser radar has a fraud attack according to the assistance detection response includes:
when a cooperative detection response is received within a specified time and the cooperative detection response contains a detection conclusion that the abnormal object does not exist at the space position at the corresponding acquisition time, confirming that the abnormal object is a false object and a spoofing attack aiming at the laser radar exists;
And when a cooperative detection response is received within a designated time, and the cooperative detection response comprises a detection conclusion that the abnormal object exists at the space position at the corresponding acquisition time, confirming that no spoofing attack exists on the laser radar, wherein the abnormal object is a real object.
In the fraud attack detection method of the embodiment of the present disclosure, the determining, according to the assistance detection response, whether the laser radar has a fraud attack, further includes:
voting a detection conclusion in a plurality of auxiliary detection responses when the plurality of auxiliary detection responses are received within a specified time;
and determining whether the laser radar has a spoofing attack according to the voting result.
In the fraud attack detection method of the embodiment of the present disclosure, after determining whether the laser radar has a fraud attack according to the assistance detection response, the method further includes:
and discarding the detection result of the false object when confirming that the abnormal object is the false object and the spoofing attack aiming at the laser radar exists.
On the other hand, the embodiment of the specification also provides another fraud attack detection method, which comprises the following steps:
receiving an assistance detection request sent by surrounding vehicles; the auxiliary detection request carries point cloud data corresponding to the abnormal object and an auxiliary detection request of a space position;
Judging whether an object which is positioned at the space position and has similarity with the abnormal object reaching a similarity threshold exists in the point cloud data of the laser radar of the vehicle at the corresponding acquisition time;
generating an assisted detection response according to the judgment result;
and returning the assistance detection response.
In the fraud attack detection method of the embodiment of the present disclosure, the generating an assisted detection response according to the determination result includes:
when an object which is positioned at the space position and has similarity with the abnormal object reaching a similarity threshold exists in point cloud data acquired at the corresponding moment by the laser radar of the vehicle, confirming that the abnormal object exists at the space position at the corresponding acquisition moment;
otherwise, confirming that the abnormal object does not exist at the space position at the corresponding acquisition time.
In the fraud attack detection method of the embodiment of the present specification, the similarity includes structural similarity.
On the other hand, the embodiment of the specification also provides a spoofing attack detecting device, which comprises:
the acquisition module is used for acquiring point cloud data currently acquired by the laser radar of the vehicle;
the identification module is used for identifying whether an abnormal object exists in the point cloud data;
The broadcasting module is used for broadcasting an auxiliary detection request carrying the point cloud data and the space position corresponding to the abnormal object to surrounding vehicles when the abnormal object exists in the point cloud data;
the receiving module is used for receiving an assistance detection response returned by the surrounding vehicles aiming at the assistance detection request;
and the determining module is used for determining whether the laser radar has a spoofing attack or not according to the assistance detection response.
On the other hand, the embodiment of the specification also provides another fraud attack detecting apparatus, including:
the receiving module is used for receiving an assistance detection request sent by surrounding vehicles; the auxiliary detection request carries point cloud data corresponding to the abnormal object and an auxiliary detection request of a space position;
the judging module is used for judging whether an object which is positioned at the space position and has similarity with the abnormal object reaching a similarity threshold exists in the point cloud data of the laser radar of the vehicle at the corresponding acquisition time;
the generation module is used for generating an assisted detection response according to the judgment result;
and the return module is used for returning the assistance detection response.
In another aspect, embodiments of the present disclosure further provide a computer device including a memory, a processor, and a computer program stored on the memory, which when executed by the processor, performs the instructions of the above method.
In another aspect, embodiments of the present disclosure also provide a computer storage medium having stored thereon a computer program which, when executed by a processor of a computer device, performs instructions of the above method.
In another aspect, embodiments of the present specification also provide a computer program product comprising a computer program which, when executed by a processor, performs instructions of the above method.
According to the technical scheme provided by the embodiment of the specification, when any automatic driving vehicle recognizes that an abnormal object exists based on the point cloud data currently collected by the laser radar of the automatic driving vehicle, an auxiliary detection request carrying the point cloud data corresponding to the abnormal object and the space position can be broadcast to surrounding vehicles, and on the basis, when an auxiliary detection response returned by the surrounding vehicles aiming at the auxiliary detection request is received, whether the laser radar of the automatic driving vehicle has a spoofing attack or not can be further verified according to the auxiliary detection response, so that the detection accuracy of the spoofing attack aiming at the laser radar is improved, and the automatic driving system can carry out more reasonable decision processing according to the accurate detection result, thereby being beneficial to improving the driving safety of the automatic driving vehicle.
Drawings
In order to more clearly illustrate the embodiments of the present description or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some of the embodiments described in the present description, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
FIG. 1 illustrates a flow chart of a spoofing attack detection method in some embodiments of the present description;
FIG. 2 illustrates a schematic diagram of using two vehicles to detect spoofing attacks in some embodiments of the present description;
FIG. 3 illustrates a flow chart for identifying whether an anomalous object exists in the point cloud data in the embodiment of FIG. 1;
FIG. 4 is a flow chart illustrating a method of spoofing attack detection in further embodiments of the present description;
FIG. 5 is a flow chart illustrating a method of spoofing attack detection in further embodiments of the present description;
FIG. 6 illustrates a block diagram of a spoofing attack detecting device in some embodiments of the present specification;
FIG. 7 is a block diagram showing the structure of a spoofing attack detecting apparatus in other embodiments of the present specification;
Fig. 8 illustrates a block diagram of a computer device in some embodiments of the present description.
[ reference numerals description ]
10. A first vehicle;
20. a second vehicle;
61. an acquisition module;
62. an identification module;
63. a broadcasting module;
64. a receiving module;
65. a determining module;
71. a receiving module;
72. a judging module;
73. a generating module;
74. a return module;
802. a computer device;
804. a processor;
806. a memory;
808. a driving mechanism;
810. an input/output interface;
812. an input device;
814. an output device;
816. a presentation device;
818. a graphical user interface;
820. a network interface;
822. a communication link;
824. a communication bus.
Detailed Description
In order to make the technical solutions in the present specification better understood by those skilled in the art, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are intended to be within the scope of the present disclosure.
The automatic driving system relies on cooperation of artificial intelligence, visual computing, radar, monitoring devices, navigation and positioning systems, etc., to automatically and safely operate the vehicle (i.e., automatically drive the vehicle) without any human initiative. Autopilot systems often employ lidar to detect the driving environment surrounding the vehicle. In recent years, some researchers have proposed a number of machine-learning models that can accurately measure the depth of an object and detect the object using collected data. However, the application of these models introduces new vulnerabilities that may compromise the safety of the autonomous vehicle. For example, since the lidar may be attacked (e.g., spoofing the lidar by a physical device at an insertion point), the data provided to the machine learning model may no longer be accurate (e.g., false object detection results are generated), which may affect the object detection accuracy of the machine learning model and thus the safety of the autonomous vehicle. Therefore, how to accurately detect the spoofing attack on the lidar has become a technical problem to be solved.
In view of this, the embodiments of the present disclosure provide a new fraud attack detection scheme, which may be implemented in a multi-vehicle scenario, i.e., by the cooperation of an autopilot system of a plurality of (e.g., two or more) autopilot vehicles, where the plurality of autopilot vehicles employ a lidar as a driving environment sensing module. When the automatic driving vehicles are closer in distance, the corresponding laser radars can acquire partially overlapped driving environment data, and then the cooperative recognition of whether the laser radars are deception attack can be carried out. Thus, the pitch condition may be set in advance. For example, the distance between the plurality of lidars may be set to be not more than a specified distance (e.g., 5 meters, 8 meters, 10 meters, etc.), and the specified distance may be set appropriately according to the actual situation when it is implemented.
It should be noted that the spoofing attack detected by the spoofing attack detecting scheme of the embodiment of the present specification may refer to: spoofing attacks using physical devices. Such an attack mode generally uses a photodiode to receive a laser beam emitted from a lidar, and after a certain time delay, returns the laser beam to a receiver of the lidar again using a transmitting device. Therefore, the effect of inserting a plurality of originally non-existing points (namely false points) at the designated positions in the normal point cloud data is achieved. The attack mode has two characteristics, namely, the number of points (0-200) of the inserted object is obviously smaller than that of the normal object. In addition, due to its attack equipment limitations, a spoof attacker typically attacks only one vehicle at a time. Therefore, when a spoofing attack is cooperatively identified by a plurality of autonomous vehicles, it can be advantageous to achieve a more accurate identification effect.
The embodiments of the present disclosure provide a fraud attack detection method that may be applied to an autopilot system side of an autopilot vehicle, and referring to fig. 1, in some embodiments, the fraud attack detection method may include the following steps:
and 101, acquiring point cloud data currently acquired by a laser radar of the vehicle.
Step 102, identifying whether an abnormal object exists in the point cloud data.
And step 103, broadcasting an auxiliary detection request carrying the point cloud data and the spatial position corresponding to the abnormal object to surrounding vehicles when the abnormal object exists in the point cloud data.
Step 104, receiving an assistance detection response returned by the surrounding vehicle for the assistance detection request.
Step 105, determining whether the laser radar has a spoofing attack according to the assistance detection response.
According to the spoofing attack detection method, when any one of the automatic driving vehicles recognizes that an abnormal object exists based on the point cloud data currently collected by the laser radar of the vehicle, an assistance detection request carrying the point cloud data corresponding to the abnormal object and the space position can be broadcast to surrounding vehicles, and on the basis, when an assistance detection response returned by the surrounding vehicles aiming at the assistance detection request is received, whether the spoofing attack exists by the laser radar of the vehicle can be further verified according to the assistance detection response, so that the detection accuracy of the spoofing attack aiming at the laser radar is improved, and the automatic driving system can perform more reasonable decision processing according to the accurate detection result, thereby being beneficial to improving the driving safety of the automatic driving vehicle.
For any one of the autonomous vehicles in an activated state (e.g., a driving state, etc.), the laser radar can acquire the point cloud data for representing the driving environment of the surrounding space of the vehicle in real time. For the autopilot system of the autopilot vehicle, the autopilot vehicle is the own vehicle, and other autopilot vehicles located around the autopilot vehicle are surrounding vehicles of the autopilot vehicle.
In some cases, one autonomous vehicle may be surrounded by one or more autonomous vehicles that meet the spacing conditions. For example, taking the embodiment shown in FIG. 2 as an example, both the first vehicle 10 and the second vehicle 20 are autonomous vehicles; for the first vehicle 10, there is an autonomous vehicle (i.e., the second vehicle 20) around it; while for the second vehicle 20 there is also an autonomous vehicle (i.e., the first vehicle 10) around it. In other cases, an autonomous vehicle may not have any autonomous vehicle around the autonomous vehicle that satisfies the distance condition, and in this case, whether a detection result for a spoofing attack of the lidar exists may be obtained directly according to the recognition result of whether an abnormal object exists in the point cloud data of the lidar; specifically, if an abnormal object exists, it is considered that there is a fraud attack against the lidar; if no abnormal object exists, the laser radar is considered to have no spoofing attack.
As shown in connection with fig. 3, in some embodiments, identifying whether an abnormal object exists in the point cloud data may include the following steps:
and step 301, inputting the point cloud data into a region candidate network, and predicting to obtain an object candidate frame set.
The region candidate network (Region Proposal Network) is a pre-trained neural network model that can generate object candidate boxes based on the point cloud data, where the object candidate boxes represent regions where objects may exist (i.e., the object candidate boxes represent features of relationships between points in the region where objects may exist in the point cloud data), are intermediate results of an object detection task, and final results of the object detection task are screened from the object candidate boxes. Each object candidate frame is used for selecting an area where an object may exist, and a typical driving environment generally includes an area where a plurality of objects (such as pedestrians, vehicles, etc.) may exist, so that a plurality of candidate frames may be obtained by inputting point cloud data into the area candidate network, thereby forming a candidate frame set.
Step 302, determining whether the point number of the candidate object in each object candidate frame in the object candidate frame set is in the point number space range.
The working principle of the laser radar is that a laser beam is emitted outwards, the laser beam is scattered and reflected after encountering an obstacle, and a part of light waves can be received by a receiver of the laser radar. The laser radar calculates the return time of the laser beam to know the distance between the obstacle and the laser beam. The laser radar continuously rotates to continuously send out laser pulses to scan surrounding objects, so that the surrounding environment can be modeled. Under such an operating principle, the near laser beam is received more, and the far laser beam returns less due to distance or scattering. Resulting in a situation where the object points closer to the lidar are denser and the points farther away are thinner. In short, due to the characteristics of the laser radar, the point cloud data has relatively dense points of objects close to the laser radar and relatively sparse points of objects far from the laser radar. The points of the object in the point cloud that appear in the lidar field of view should follow a distribution law that varies from large to small with distance from near to far under normal conditions (e.g., without occlusion and spoofing attacks).
Meanwhile, as the angles of objects in the laser radar field facing the laser radar are different, the scanned surfaces are different, and the points of the same object at the same distance can have different values, the maximum value and the minimum value of the points of the object at the same distance from the laser radar can be modeled in advance according to the angles. The method can be used for drawing a regression curve of the minimum value of the point number of the object in the laser radar field along with the change of the distance and a regression curve of the maximum value of the point number of the object in the laser radar field along with the change of the distance in advance, and the space between the two regression curves is a preset point number space range (namely a normal point number space range). Research shows that the number of points of false objects inserted into the point cloud of the attacked vehicle is almost sparse, and does not accord with the due points of the true objects at the positions of the points, namely does not accord with the distribution rule, so that the laser radar can be favorably identified whether the laser radar is deceptively attacked or not through the preset point space range.
In some embodiments, the two regression curves may be modeled using an autopilot simulator (e.g., carla, etc.) simulation. At the same time, since the lidar parameters in the autopilot simulator are adjustable (e.g., rotational frequency, number of laser beams emitted per second, etc.). By adopting the method, regression curves of the laser radars of different types (namely, the point space ranges of the laser radars of different types) can be obtained offline, so that the spoofing attack detection method of the embodiment of the specification has stronger applicability.
In most cases, there may be multiple object candidate boxes within the object candidate box set; when the object candidate frame is assembled with a plurality of object candidate frames, each object candidate frame needs to respectively confirm whether the points of the candidate objects in the object candidate frame are in the point space range.
The point cloud data collected by the laser radar is a massive point set of object surface characteristics, and for the candidate object in each object candidate frame, the object surface of the candidate object is also formed by combining a plurality of points, and the corresponding number of the points is the number of points of the candidate object.
Step 303, when the number of points of the candidate object in the object candidate frame is not in the number space range, judging whether other object candidate frames exist in the distance range from the candidate object to the laser radar.
And step 304, if no other object candidate frame exists in the distance range from the object candidate to the laser radar, confirming that the object is an abnormal object.
Judging whether other object candidate frames exist in the distance range from the candidate object to the laser radar or not, wherein the method can be used for judging whether the candidate object is shielded or not; if other object candidate frames exist in the distance range between the object candidate and the laser radar, the point corresponding to the object candidate is indicated to be less than the normal point space range due to occlusion (namely, the object candidate frames are occluded by the object candidate in the other object candidate frames); otherwise, the candidate object may be considered as an abnormal object. The abnormal object has a high probability of being a false object. This is sometimes referred to herein as an anomalous object before further confirmation.
In order to improve accuracy of fraud attack detection, when an abnormal object exists in the point cloud data, an assistance detection request carrying the point cloud data and the spatial position corresponding to the abnormal object may be broadcast to surrounding vehicles to request surrounding vehicle assistance detection (assistance processing logic of surrounding other vehicles will be described in detail below). In other embodiments, the manner of sending the request for assistance detection may be replaced by other communication manners such as multicast or peer-to-peer, as desired.
In an autopilot scenario, real-time is extremely important, so a short assistance detection response waiting time can be preset, and if an assistance detection response for the assistance detection request is received within the assistance detection response waiting time, whether a spoofing attack exists in the lidar can be determined accordingly. Furthermore, the detection conclusion in the assisted detection response may be: the abnormal object exists at the spatial position at the corresponding acquisition time, or the abnormal object does not exist at the spatial position at the corresponding acquisition time.
As already explained above, due to their attack equipment limitations, spoof attackers typically attack only one vehicle at a time. Thus, when an autonomous vehicle finds that an abnormal object exists at a certain spatial location at a certain acquisition time:
if the abnormal object is an abnormal object inserted by spoofing attack, surrounding vehicles cannot detect the abnormal object at the space position corresponding to the acquisition time; therefore, when a cooperative detection response is received within a specified time and the cooperative detection response includes a detection conclusion that the abnormal object does not exist at the spatial position at the corresponding acquisition time, the abnormal object can be confirmed to be a false object, and a spoofing attack for the laser radar exists.
If the abnormal object is a real object, surrounding vehicles can detect the abnormal object at the space position corresponding to the acquisition time; therefore, when a cooperative detection response is received within a specified time and the cooperative detection response includes a detection conclusion that the abnormal object exists at the spatial position at the corresponding acquisition time, it can be confirmed that no spoofing attack exists in the laser radar, that is, the abnormal object is a real object.
Thus, determining whether a fraud attack exists on the lidar based on the assistance detection response may include the following:
(1) When a cooperative detection response is received within a specified time (i.e., the above-mentioned waiting time of the cooperative detection response), and the cooperative detection response includes a detection conclusion that the abnormal object does not exist at the spatial position at the corresponding acquisition time, it may be confirmed that the abnormal object is a false object, and a spoofing attack for the lidar exists.
(2) When a cooperative detection response is received within a specified time and the cooperative detection response contains a detection conclusion that the abnormal object exists at the space position at the corresponding acquisition time, the laser radar can be confirmed to be free from spoofing attack, namely the abnormal object is a real object.
(3) Voting a detection conclusion in a plurality of auxiliary detection responses when the plurality of auxiliary detection responses are received within a specified time; and determining whether the laser radar has a spoofing attack according to the voting result. When multiple assisted detection responses are received within a specified time, there may be cases where the detection conclusions of the multiple assisted detection responses do not coincide exactly, so that the final detection conclusions may be decided by way of a line voting. In some embodiments, the final detection conclusion may be decided in a few majority-compliant voting fashion.
For example, if vehicle a receives multiple assistance detection responses within a specified time: an assistance detection response X returned by vehicle B, an assistance detection response Y returned by vehicle C, and an assistance detection response Z returned by vehicle D. If the detection conclusion of the auxiliary detection response X and the auxiliary detection response Y is: the abnormal object exists at the space position corresponding to the acquisition time, and the detection conclusion of the assisted detection response Z is as follows: the abnormal object does not exist at the space position corresponding to the acquisition time; the number of votes in the detection conclusion that the abnormal object exists at the spatial position corresponding to the acquisition time is 2, the number of votes in the detection conclusion that the abnormal object does not exist at the spatial position corresponding to the acquisition time is 1, and 2>1 is obvious, so that the final detection conclusion can be confirmed that the abnormal object exists at the spatial position corresponding to the acquisition time, the abnormal object can be confirmed to be a real object, and the corresponding laser radar does not have fraud attack.
(4) If no assistance detection response is received within a specified time, the abnormal object can be directly confirmed as a false object, and a spoofing attack for the laser radar exists.
The embodiments of the present disclosure provide another fraud attack detection method, which may be applied to an autopilot system side of an autopilot vehicle, and referring to fig. 4, in some embodiments, the fraud attack detection method may include the steps of:
step 401, acquiring point cloud data currently acquired by a laser radar of a vehicle.
Step 402, identifying whether an abnormal object exists in the point cloud data.
Step 403, broadcasting an assistance detection request carrying the point cloud data and the spatial position corresponding to the abnormal object to surrounding vehicles when the abnormal object exists in the point cloud data.
Step 404, receiving an assistance detection response returned by the surrounding vehicle for the assistance detection request.
Step 405, determining whether the laser radar has a spoofing attack according to the assistance detection response.
Step 406, discarding the detection result of the false object when confirming that the abnormal object is the false object and the spoofing attack aiming at the laser radar exists.
When the abnormal object is confirmed to be a false object and the spoofing attack to the laser radar exists, the false object detection result is discarded, so that sudden braking decision caused by the false object being considered as a real object can be avoided, the driving safety of an automatic driving vehicle is improved, and the energy consumption increase caused by unnecessary sudden braking is avoided. The detection result of the false object is discarded, for example, the detection result of the false object is deleted or ignored.
In the above embodiments of the fraud attack detection method, the automatic driving system of the automatic driving vehicle sending the assistance detection request is used as the execution subject, and in order to facilitate a clearer and complete understanding of the processing logic of the whole fraud attack detection, another fraud attack detection method is described below using the automatic driving system of the automatic driving vehicle returning the assistance detection response as the execution subject. Referring to fig. 5, in some embodiments, the spoofing attack detecting method may include the steps of:
step 501, receiving an assistance detection request sent by surrounding vehicles; the auxiliary detection request carries point cloud data corresponding to the abnormal object and a space position auxiliary detection request.
In consideration of dynamic changes of driving environments, point cloud data collected by the laser radar at different sampling moments are generally different, so that corresponding collection time stamps can be attached to point cloud data and space positions carried in the assistance detection request.
Step 502, judging whether an object which is positioned at the space position and has similarity with the abnormal object reaching a similarity threshold exists in the point cloud data corresponding to the acquisition time of the laser radar of the vehicle.
The host vehicle herein is with respect to an autonomous vehicle that receives a request for assistance in detection. For each autonomous vehicle that receives the assistance detection request, it may be determined whether an object that is located at the spatial position and has a similarity with the abnormal object that reaches a similarity threshold value exists in the point cloud data of the laser radar of the vehicle at the corresponding acquisition time according to the following manner:
(1) And finding out the point cloud data of the laser radar of the vehicle at the corresponding acquisition time according to the acquisition time stamp.
(2) And inputting point cloud data of the laser radar of the vehicle at the corresponding acquisition time into the area candidate network, and predicting to obtain an object candidate frame set.
(3) Whether an object candidate frame corresponding to the spatial position in the assistance detection request exists in the object candidate frame set is judged. If an object candidate frame corresponding to the spatial position in the auxiliary detection request exists, the next similarity judgment is carried out, otherwise, the existence of the abnormal object at the spatial position at the corresponding acquisition time can be confirmed.
(4) When an object candidate frame corresponding to the space position in the auxiliary detection request exists, comparing the similarity between the candidate object of the object candidate frame and the abnormal object in the auxiliary detection request; if the similarity reaches a similarity threshold, confirming that the abnormal object exists at the space position at the corresponding acquisition time; otherwise it can be considered that no abnormal object is present at the spatial position at the corresponding acquisition instant. In some embodiments, the similarity comparison may be implemented, for example, using a structural similarity (Structure Similarity, SSIM) algorithm.
And step 503, generating an assisted detection response according to the judgment result.
Step 504, returning the assistance detection response.
In this way, the automatic driving vehicle sending the assistance detection request can further judge whether the abnormal object is a false object by means of the assistance detection response, so that the accuracy of the fraud attack detection is improved.
While the process flows described above include a plurality of operations occurring in a particular order, it should be apparent that the processes may include more or fewer operations, which may be performed sequentially or in parallel (e.g., using a parallel processor or a multi-threaded environment).
Corresponding to the above-mentioned fraud attack detection method shown in fig. 1, the embodiment of the present disclosure further provides a fraud attack detection apparatus, which may be configured on the above-mentioned autopilot system, and referring to fig. 6, the fraud attack detection apparatus may include:
the acquiring module 61 may be configured to acquire point cloud data currently acquired by a lidar of the vehicle;
an identifying module 62, configured to identify whether an abnormal object exists in the point cloud data;
a broadcasting module 63, configured to, when an abnormal object exists in the point cloud data, broadcast an assistance detection request carrying the point cloud data and a spatial position corresponding to the abnormal object to surrounding vehicles;
a receiving module 64, configured to receive an assistance detection response returned by the surrounding vehicle for the assistance detection request;
a determining module 65 may be configured to determine whether a fraud attack exists on the lidar based on the assistance detection response.
In some embodiments of the fraud attack detecting apparatus, the identifying whether an abnormal object exists in the point cloud data includes:
inputting the point cloud data into a region candidate network, and predicting to obtain an object candidate frame set;
Determining whether the point number of the candidate object in each object candidate frame in the object candidate frame set is positioned in a point number space range;
when the number of points of the candidate object in the object candidate frame is not in the number space range, judging whether other object candidate frames exist in the distance range from the candidate object to the laser radar;
and if no other object candidate frame exists in the distance range from the object candidate to the laser radar, confirming that the object is an abnormal object.
In the fraud attack detecting apparatus of some embodiments, the point space range includes a point space between a first regression curve and a second regression curve;
the first regression curve is a regression curve of the minimum value of the point number of the object in the laser radar field of view along with the change of the detection distance under the normal condition;
and the second regression curve is a regression curve of the maximum value of the point number of the object in the laser radar field of view along with the change of the detection distance under the normal condition.
In the fraud attack detection apparatus of some embodiments, the determining whether the laser radar has a fraud attack according to the assistance detection response includes:
when a cooperative detection response is received within a specified time and the cooperative detection response contains a detection conclusion that the abnormal object does not exist at the space position at the corresponding acquisition time, confirming that the abnormal object is a false object and a spoofing attack aiming at the laser radar exists;
And when a cooperative detection response is received within a designated time, and the cooperative detection response comprises a detection conclusion that the abnormal object exists at the space position at the corresponding acquisition time, confirming that no spoofing attack exists on the laser radar, wherein the abnormal object is a real object.
In the fraud attack detection apparatus of some embodiments, the determining whether the laser radar has a fraud attack according to the assistance detection response further includes:
voting a detection conclusion in a plurality of auxiliary detection responses when the plurality of auxiliary detection responses are received within a specified time;
and determining whether the laser radar has a spoofing attack according to the voting result.
The spoofing attack detecting device of some embodiments further includes a decision module, where the decision module is configured to discard a detection result of the false object when the abnormal object is confirmed to be a false object and a spoofing attack for the laser radar exists after the determination module determines whether the laser radar has the spoofing attack according to the assisted detection response.
Corresponding to the above-mentioned fraud attack detection method shown in fig. 5, another fraud attack detection apparatus is provided in the embodiment of the present disclosure, which may be configured on the above-mentioned autopilot system, and referring to fig. 7, the fraud attack detection apparatus may include:
A receiving module 71, which may be configured to receive an assistance detection request sent by a surrounding vehicle; the auxiliary detection request carries point cloud data corresponding to the abnormal object and an auxiliary detection request of a space position;
the judging module 72 may be configured to judge whether an object that is located at the spatial position and has a similarity with the abnormal object that reaches a similarity threshold exists in the point cloud data corresponding to the acquisition time of the laser radar of the host vehicle;
a generating module 73, configured to generate an assisted detection response according to the determination result;
a return module 74 may be used to return the assistance detection response.
In the fraud attack detecting apparatus of some embodiments, the generating an assisted detection response according to the determination result includes:
when an object which is positioned at the space position and has similarity with the abnormal object reaching a similarity threshold exists in point cloud data acquired at the corresponding moment by the laser radar of the vehicle, confirming that the abnormal object exists at the space position at the corresponding acquisition moment;
otherwise, confirming that the abnormal object does not exist at the space position at the corresponding acquisition time.
In the fraud attack detecting apparatus of some embodiments, the similarity includes a structural similarity.
For convenience of description, the above devices are described as being functionally divided into various units, respectively. Of course, the functions of each element may be implemented in one or more software and/or hardware elements when implemented in the present specification.
Embodiments of the present description also provide a computer device. As shown in fig. 8, in some embodiments of the present description, the computer device 802 may include one or more processors 804, such as one or more Central Processing Units (CPUs) or Graphics Processors (GPUs), each of which may implement one or more hardware threads. The computer device 802 may also include any memory 806 for storing any kind of information, such as code, settings, data, etc., and in a particular embodiment, a computer program on the memory 806 and executable on the processor 804, which when executed by the processor 804, may perform the instructions of the spoofing attack detection method described in any of the embodiments above. For example, and without limitation, memory 806 may include any one or more of the following combinations: any type of RAM, any type of ROM, flash memory devices, hard disks, optical disks, etc. More generally, any memory may store information using any technique. Further, any memory may provide volatile or non-volatile retention of information. Further, any memory may represent fixed or removable components of computer device 802. In one case, the computer device 802 may perform any of the operations of the associated instructions when the processor 804 executes the associated instructions stored in any memory or combination of memories. The computer device 802 also includes one or more drive mechanisms 808, such as a hard disk drive mechanism, an optical disk drive mechanism, and the like, for interacting with any memory.
The computer device 802 may also include an input/output interface 810 (I/O) for receiving various inputs (via an input device 812) and for providing various outputs (via an output device 814). One particular output mechanism may include a presentation device 816 and an associated graphical user interface 818 (GUI). In other embodiments, input/output interface 810 (I/O), input device 812, and output device 814 may not be included, but merely as a computer device in a network. The computer device 802 may also include one or more network interfaces 820 for exchanging data with other devices via one or more communication links 822. One or more communications buses 824 couple the above-described components together.
The communication link 822 may be implemented in any manner, such as, for example, through a local area network, a wide area network (e.g., the internet), a point-to-point connection, etc., or any combination thereof. Communication link 822 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc., governed by any protocol or combination of protocols.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), computer-readable storage media and computer program products according to some embodiments of the specification. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations 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 processor to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processor, 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 processor 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 processor 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 computer device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
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 storage media for a computer 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 Disks (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computer device. Computer readable media, as defined in the specification, does not include transitory computer readable media (transmission media), such as modulated data signals and carrier waves.
It will be appreciated by those skilled in the art that embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, the present specification embodiments may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present description embodiments may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
The present embodiments 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 embodiments of the specification may also be practiced in distributed computing environments where tasks are performed by remote processors 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.
It should also be understood that, in the embodiments of the present specification, the term "and/or" is merely one association relationship describing the association object, meaning that three relationships may exist. For example, a and/or B may represent: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the embodiments of the present specification. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.
Claims (14)
1. A fraud attack detection method, comprising:
acquiring point cloud data currently acquired by a laser radar of the vehicle;
identifying whether an abnormal object exists in the point cloud data;
broadcasting an auxiliary detection request carrying point cloud data and a space position corresponding to the abnormal object to surrounding vehicles when the abnormal object exists in the point cloud data;
receiving an assistance detection response returned by the surrounding vehicles aiming at the assistance detection request;
and determining whether the laser radar has a spoofing attack according to the assistance detection response.
2. The fraud attack detection method of claim 1, wherein the identifying whether an abnormal object exists in the point cloud data includes:
inputting the point cloud data into a region candidate network, and predicting to obtain an object candidate frame set;
determining whether the point number of the candidate object in each object candidate frame in the object candidate frame set is positioned in a point number space range;
when the number of points of the candidate object in the object candidate frame is not in the number space range, judging whether other object candidate frames exist in the distance range from the candidate object to the laser radar;
And if no other object candidate frame exists in the distance range from the object candidate to the laser radar, confirming that the object is an abnormal object.
3. The fraud attack detection method of claim 2, wherein the point space range includes a point space between the first regression curve and the second regression curve;
the first regression curve is a regression curve of the minimum value of the point number of the object in the laser radar field of view along with the change of the detection distance under the normal condition;
and the second regression curve is a regression curve of the maximum value of the point number of the object in the laser radar field of view along with the change of the detection distance under the normal condition.
4. The fraud attack detection method of claim 1, wherein the determining whether the lidar is in a fraud attack based on the assistance detection response includes:
when a cooperative detection response is received within a specified time and the cooperative detection response contains a detection conclusion that the abnormal object does not exist at the space position at the corresponding acquisition time, confirming that the abnormal object is a false object and a spoofing attack aiming at the laser radar exists;
and when a cooperative detection response is received within a designated time, and the cooperative detection response comprises a detection conclusion that the abnormal object exists at the space position at the corresponding acquisition time, confirming that no spoofing attack exists on the laser radar, wherein the abnormal object is a real object.
5. The fraud attack detection method of claim 4, wherein the determining whether the lidar is subject to a fraud attack based on the assistance detection response further comprises:
voting a detection conclusion in a plurality of auxiliary detection responses when the plurality of auxiliary detection responses are received within a specified time;
and determining whether the laser radar has a spoofing attack according to the voting result.
6. The fraud attack detection method of claim 4, further comprising, after determining from the assistance detection response whether a fraud attack is present by the lidar:
and discarding the detection result of the false object when confirming that the abnormal object is the false object and the spoofing attack aiming at the laser radar exists.
7. A fraud attack detection method, comprising:
receiving an assistance detection request sent by surrounding vehicles; the auxiliary detection request carries point cloud data corresponding to the abnormal object and an auxiliary detection request of a space position;
judging whether an object which is positioned at the space position and has similarity with the abnormal object reaching a similarity threshold exists in the point cloud data of the laser radar of the vehicle at the corresponding acquisition time;
Generating an assisted detection response according to the judgment result;
and returning the assistance detection response.
8. The fraud attack detection method of claim 7, wherein generating the assistance detection response according to the determination result includes:
when an object which is positioned at the space position and has similarity with the abnormal object reaching a similarity threshold exists in point cloud data acquired at the corresponding moment by the laser radar of the vehicle, confirming that the abnormal object exists at the space position at the corresponding acquisition moment;
otherwise, confirming that the abnormal object does not exist at the space position at the corresponding acquisition time.
9. The spoofing attack detecting method recited in claim 7 wherein the similarity comprises a structural similarity.
10. A fraud attack detecting apparatus, comprising:
the acquisition module is used for acquiring point cloud data currently acquired by the laser radar of the vehicle;
the identification module is used for identifying whether an abnormal object exists in the point cloud data;
the broadcasting module is used for broadcasting an auxiliary detection request carrying the point cloud data and the space position corresponding to the abnormal object to surrounding vehicles when the abnormal object exists in the point cloud data;
The receiving module is used for receiving an assistance detection response returned by the surrounding vehicles aiming at the assistance detection request;
and the determining module is used for determining whether the laser radar has a spoofing attack or not according to the assistance detection response.
11. A fraud attack detecting apparatus, comprising:
the receiving module is used for receiving an assistance detection request sent by surrounding vehicles; the auxiliary detection request carries point cloud data corresponding to the abnormal object and an auxiliary detection request of a space position;
the judging module is used for judging whether an object which is positioned at the space position and has similarity with the abnormal object reaching a similarity threshold exists in the point cloud data of the laser radar of the vehicle at the corresponding acquisition time;
the generation module is used for generating an assisted detection response according to the judgment result;
and the return module is used for returning the assistance detection response.
12. A computer device comprising a memory, a processor, and a computer program stored on the memory, characterized in that the computer program, when being executed by the processor, performs the instructions of the method according to any one of claims 1-9.
13. A computer storage medium having stored thereon a computer program, which, when executed by a processor of a computer device, performs the instructions of the method according to any of claims 1-9.
14. A computer program product, characterized in that the computer program product comprises a computer program which, when being executed by a processor, executes instructions of the method according to any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210567504.6A CN117152708A (en) | 2022-05-24 | 2022-05-24 | Spoofing attack detection method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210567504.6A CN117152708A (en) | 2022-05-24 | 2022-05-24 | Spoofing attack detection method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117152708A true CN117152708A (en) | 2023-12-01 |
Family
ID=88897406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210567504.6A Pending CN117152708A (en) | 2022-05-24 | 2022-05-24 | Spoofing attack detection method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117152708A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118151543A (en) * | 2024-05-11 | 2024-06-07 | 北京航空航天大学杭州创新研究院 | Unmanned vehicle cooperative control method and device |
-
2022
- 2022-05-24 CN CN202210567504.6A patent/CN117152708A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118151543A (en) * | 2024-05-11 | 2024-06-07 | 北京航空航天大学杭州创新研究院 | Unmanned vehicle cooperative control method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP2021523443A (en) | Association of lidar data and image data | |
| EP3602396A1 (en) | Embedded automotive perception with machine learning classification of sensor data | |
| US11450120B2 (en) | Object detection in point clouds | |
| US12073329B2 (en) | Method for recognizing an adversarial disturbance in input data of a neural network | |
| CN111783905B (en) | Target fusion method and device, storage medium and electronic equipment | |
| CN110261856A (en) | A kind of radar detection method and device based on the collaboration detection of more radars | |
| US12084051B2 (en) | Scenario identification in autonomous driving environments | |
| EP3703008A1 (en) | Object detection and 3d box fitting | |
| CN111369617A (en) | 3D target detection method of monocular view based on convolutional neural network | |
| CN111695619A (en) | Multi-sensor target fusion method and device, vehicle and storage medium | |
| CN117152708A (en) | Spoofing attack detection method, device, equipment and storage medium | |
| CN115147333A (en) | Target detection method and device | |
| CN111352111B (en) | Positioning and/or classifying objects | |
| CN109263649B (en) | Vehicle, object recognition method and object recognition system thereof in automatic driving mode | |
| EP3975042B1 (en) | Method and apparatus for determining running region information | |
| CN113900101A (en) | Obstacle detection method and device and electronic equipment | |
| TWI735191B (en) | System and method for lidar defogging | |
| JP2019164098A (en) | Road surface shape estimation device, road surface shape estimation program, and moving body | |
| CN116612638A (en) | Traffic collision accident detection method, device and readable medium | |
| Shanshan et al. | An evaluation system based on user big data management and artificial intelligence for automatic vehicles | |
| EP4099211A1 (en) | Method and device for training a machine learning algorithm | |
| US20220065989A1 (en) | Augmentation and processing of radar data with machine learning | |
| CN115407281A (en) | Point cloud denoising method and device, detection radar and storage medium | |
| CN114973171A (en) | Preventing low speed edge-wiping collision with non-moving objects | |
| CN108885262A (en) | For capturing the method for at least one object, the device of sensor device, sensor device and with the driver assistance system of at least one sensor device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |