WO2019119350A1 - 用于无人驾驶车辆的障碍物识别方法、装置及终端设备 - Google Patents

用于无人驾驶车辆的障碍物识别方法、装置及终端设备 Download PDF

Info

Publication number
WO2019119350A1
WO2019119350A1 PCT/CN2017/117730 CN2017117730W WO2019119350A1 WO 2019119350 A1 WO2019119350 A1 WO 2019119350A1 CN 2017117730 W CN2017117730 W CN 2017117730W WO 2019119350 A1 WO2019119350 A1 WO 2019119350A1
Authority
WO
WIPO (PCT)
Prior art keywords
laser
obstacle
unmanned vehicle
laser signal
static
Prior art date
Application number
PCT/CN2017/117730
Other languages
English (en)
French (fr)
Inventor
黄佳敏
闫文境
黄骏
Original Assignee
深圳市海梁科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳市海梁科技有限公司 filed Critical 深圳市海梁科技有限公司
Publication of WO2019119350A1 publication Critical patent/WO2019119350A1/zh

Links

Classifications

    • 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
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/88—Lidar systems specially adapted for specific applications
    • G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
    • G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • 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
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/86—Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders

Definitions

  • the present application belongs to the technical field of vehicles, and in particular, to an obstacle recognition method, a terminal device, and a medium for an unmanned vehicle.
  • a driverless car is a smart car that senses the road environment through an in-vehicle sensing system, automatically plans driving routes, and controls the vehicle to reach a predetermined target. Because it can sense road information, vehicle position and obstacle information, it controls the steering and speed of the vehicle, thus ensuring that the driverless car can travel safely and reliably on the road.
  • unmanned vehicles on the market generally detect obstacles by real-time three-dimensional rotary laser radar.
  • Such laser radars sense the environment in a dynamic scanning manner.
  • the transmitting/receiving component inside the laser radar is combined with the rotating mirror surface, and by rotating the mirror surface, a viewing angle of 90 to 180 degrees can be achieved.
  • the mirror not only reflects the light emitted by the diode, but also reflects the reflected light back to the receiver, so that the photon detector mounted near the diode can detect the returned signal to collect a large number of point clouds. In turn, a three-dimensional image of the surrounding environment is generated.
  • the update frequency can reach 10 Hz, but in the vertical direction, the angle of view is still small, about 20 degrees, so that even high-speed rotation is performed in the horizontal dimension. It scans 10 times per second and emits 2000 laser points, but in the final point cloud, the vertical point resolution is also sparse. Therefore, when the obstacle is identified based on the point cloud, there may still be some scanning blind spots, for example, scanning defects such as some relatively small remote poles or near fences, thereby reducing the accuracy of obstacle detection. When an obstacle occurs in the actual environment, the unmanned vehicle still determines that the current obstacle is present and continues to advance, thereby reducing the safety performance of the unmanned vehicle.
  • the embodiments of the present application provide an obstacle recognition method, device, terminal device and medium for an unmanned vehicle, so as to solve the problem that the obstacle detection accuracy and the safety performance of the currently unmanned vehicle are relatively low. The problem.
  • a first aspect of the embodiments of the present application provides an obstacle recognition method for an unmanned vehicle, including:
  • the obstacle information is transmitted to the vehicle controller to cause the vehicle controller to determine an operating state of the driverless vehicle based on the obstacle information.
  • a second aspect of the embodiments of the present application provides an obstacle recognition apparatus for an unmanned vehicle, including:
  • control unit configured to control a plurality of static laser radars on the unmanned vehicle to respectively transmit a first laser signal, wherein the plurality of static laser radars are respectively located around the body of the unmanned vehicle;
  • a receiving unit configured to receive a second laser signal obtained after the first laser signal is reflected
  • An acquiring unit configured to perform identification processing on the second laser signal received in each direction to acquire obstacle information
  • a sending unit configured to send the obstacle information to the vehicle controller, so that the vehicle controller determines an operating state of the unmanned vehicle based on the obstacle information.
  • a third aspect of the embodiments of the present application provides a terminal device, including a memory, a processor, and a computer program stored on the memory and operable on the processor, the processor executing the computer program
  • the step of the obstacle recognition method for an unmanned vehicle according to the first aspect is achieved.
  • a fourth aspect of the embodiments of the present application provides a computer readable storage medium storing a computer program, when the computer program is executed by at least one processor, implementing the method as described in the first aspect The steps of the obstacle recognition method for an unmanned vehicle.
  • the omnidirectional laser scanning of the obstacles is performed in a fixed position manner. Since the static lidar does not need to perform three-dimensional rotary operation, but only needs fixed-point detection, it avoids the occurrence of scan omission and improves the recognition accuracy of obstacles. Secondly, it is because static laser radar does not need dynamic rotation. The operation reduces the burden on the driverless vehicle, for example, reduces the heating energy consumption of the vehicle body, and also prolongs the service life of the laser radar, thereby making it more stable. In the third aspect, compared with the rotary laser radar, the cost of the static laser radar is greatly reduced, so that the obstacle recognition accuracy is improved, and the manufacturing cost of the unmanned vehicle is also indirectly reduced.
  • FIG. 1 is a flowchart of an implementation of an obstacle recognition method for an unmanned vehicle provided by an embodiment of the present application
  • FIG. 2 is a top view of an unmanned bus provided with static laser radar around the vehicle body according to an embodiment of the present application;
  • FIG. 3 is an internal structural diagram of a static laser radar provided by an embodiment of the present application.
  • FIG. 5 is a flowchart of an implementation of an obstacle recognition method for an unmanned vehicle according to another embodiment of the present application.
  • FIG. 6 is a top view of a driverless bus provided with static laser radar on both sides of a vehicle door according to another embodiment of the present application;
  • FIG. 7 is a structural block diagram of an obstacle recognition apparatus for an unmanned vehicle according to an embodiment of the present application.
  • FIG. 8 is a schematic diagram of a terminal device according to an embodiment of the present application.
  • FIG. 1 shows an implementation flow of an obstacle recognition method for an unmanned vehicle provided by an embodiment of the present application, and the method flow includes steps S101 to S104.
  • the specific implementation principle of each step is as follows.
  • S101 Control a plurality of static laser radars on the unmanned vehicle to respectively emit a first laser signal, and the plurality of static laser radars are respectively located around the body of the unmanned vehicle.
  • the static laser radar is a non-rotating laser radar, and may be, for example, a micro-electromechanical system (MEMS) micro-motion laser radar or the like.
  • MEMS micro-electromechanical system
  • the static lidar is located around the body of the driverless vehicle such that at least one static lidar is present at each corner of the driverless vehicle.
  • the above-described unmanned vehicles include, but are not limited to, unmanned buses and unmanned cars.
  • FIG. 2 is a top view of a driverless bus with static laser radar installed around the vehicle body according to an embodiment of the present application.
  • the left and right sides of the front and rear sides of the driverless bus are respectively provided with a static laser radar 3, which can be located at the four corners of the vehicle body or on each side of the vehicle body. The intersection.
  • the angle between the laser signal emitted by the static lidar and the extension of the body length is fixed.
  • each of the static laser radars when receiving the running instruction of the unmanned vehicle, each of the static laser radars is activated to control the laser transmitter inside each static laser radar to emit a laser signal. Since the position angles of the static laser radars are different, for the entire unmanned vehicle, laser signals of various directions can be emitted, so that the scanning position of the laser signals can cover the various oblique directions that may affect the driving of the unmanned vehicle. region.
  • the laser light source of each static laser radar on the unmanned vehicle is controlled to be reflected by the convex mirror to form a laser signal, thereby causing the laser signal to undergo specular reflection. It is then projected onto the target object of the outside world.
  • the convex mirror is located inside the static laser radar.
  • FIG. 3 is a schematic diagram of the internal structure of a static laser radar according to an embodiment of the present application.
  • the laser emitter 31 inside the static laser radar acts as a laser light source, and emits a laser beam carrying the above laser signal.
  • the convex mirror 32 the direction toward which the convex side faces is opposite to the direction in which the laser beam is emitted.
  • the laser lines included in the laser beam are projected at different positions on the convex mirror 32, and are located at various positions. After the specular reflection is emitted, the target object is projected outside the static lidar.
  • a convex mirror opposite to the laser beam emitting direction is disposed inside the static laser radar, so that the laser signal emitted by the laser light source can be emitted through the convex surface of the convex mirror and then projected to the external environment.
  • the static laser radar the laser signal is directly projected to the external environment, and the scanning area of the static laser radar is enlarged, so that the laser signal emitted by the static laser radar at a fixed position can cover a wider target object. In order to achieve a more comprehensive obstacle detection effect, the obstacle detection efficiency is improved.
  • S102 Receive a second laser signal obtained after the first laser signal is reflected.
  • the laser emitter inside the static lidar acts as a laser source, and the pulsed laser signal emitted will hit the scanning area of the static lidar. If there are obstacles in the scanning area, for example, when there are trees, roads, bridges or buildings, the pulsed laser signal will be scattered. At this time, part of the laser signal will be reflected to the receiver inside the static laser radar.
  • the laser signal reflected by the reflected laser wave is received by the built-in receiver of the static laser radar.
  • the unmanned vehicle since the driverless vehicle needs to correctly recognize the traveling road, the unmanned vehicle is prevented from traveling to the pedestrian road. Therefore, in the side of the sidewalk, a reflective strip is attached, and the above S102 specifically includes:
  • the above-mentioned pedestrian road refers to other roads located on both sides of the road of the vehicle and higher than the road on which the vehicle travels; the side of the pedestrian road is the plane where the height line of the pedestrian road is located.
  • the static laser radar is preset around the body of the unmanned vehicle, and the height of the static laser radar on the left and right sides of the vehicle head and the emission angle of the laser signal are adjusted so that each static laser radar on the left and right sides of the vehicle head It can clearly and completely scan the diffuse strip on the pedestrian road to ensure that the laser signal from the left and right sides of the front of the vehicle can completely cover the diffuse strip during the operation, regardless of the position of the driver's road on the road. The minimum angle.
  • the height of the static lidar on the left and right sides of the rear and the launch angle of the laser signal are adjusted, so that each static laser radar on the left and right sides of the rear can be scanned with a higher probability.
  • Driving another vehicle behind the vehicle can further perform a safe lane change operation without detecting the rear vehicle.
  • the surface of the diffuse reflection strip is covered with a matte material, and the matte material is used for diffusely reflecting a laser signal of 905 nm.
  • the matte material used for the diffuse reflective strip is a mixture of titanium dioxide R-930, barium sulfate, and glass microbeads.
  • barium sulfate is used as a dispersing agent
  • glass beads are used as a filler.
  • the glass microbead refers to a glass bead obtained by using glass sand as a raw material and subjecting the raw material to a high temperature melting treatment.
  • the glass beads have a diameter between 75 microns and 1400 microns and a refractive index between 1.50 and 1.64.
  • the wavelength and particle diameter of the scattered light satisfy the following relationship:
  • nR is the resin refractive index
  • m is the scattering ability of the matte material
  • k is the preset constant associated with m and nR
  • d is the particle size of the matte material
  • ⁇ is the wavelength of the scattered laser signal .
  • the titanium dioxide can be uniformly dispersed during the mixing process, the diffuse reflection ability of the diffuse reflection strip is improved, and the cost of the matte material is reduced.
  • the diffuse reflection ability and the retroreflective ability of the diffuse strip are also improved to some extent. Since the side of the pedestrian road is provided with a diffuse reflection strip, the diffuse reflection effect of the diffuse reflection strip enables the laser signal emitted by the static lidar to be reflected by the diffuse reflection strip to obtain a more accurate and significant obstacle recognition effect. It also improves the recognition accuracy of obstacles, thereby improving the safety performance of unmanned vehicles.
  • S103 Perform identification processing on the second laser signal received in each direction to acquire obstacle information.
  • the point cloud information corresponding to the laser signal is obtained according to the laser reflection intensity and the three-dimensional coordinate data thereof. Since the point cloud is a massive point cloud data that expresses the spatial distribution of the obstacle and the surface characteristics of the obstacle in the same spatial coordinate system, the obstacle can be obtained by the laser signal of the echo received by the static laser radar at multiple positions. status information.
  • the above state information includes, but is not limited to, the distance, azimuth, altitude, speed, posture, shape, and the like of the obstacle relative to the unmanned vehicle, for example, the orientation and distance of the pedestrian road relative to the unmanned vehicle. Among them, the distance information of the obstacle is calculated by the principle of laser ranging.
  • S104 Send the obstacle information to the vehicle controller, so that the vehicle controller determines an operating state of the unmanned vehicle based on the obstacle information.
  • the obstacle information updated in real time is sent to the vehicle controller.
  • the vehicle controller is used to determine the running state of the unmanned vehicle, and the running state includes the traveling speed, the steering information of each wheel, and the emergency brake information.
  • the obstacle information acquired by the vehicle controller is that there is an obstacle at a distance of 1 meter ahead, according to a preset obstacle avoidance algorithm, it can be determined that the running state of the unmanned vehicle is an emergency brake;
  • the object information is specifically that the direction in which the straight line of the vehicle body extends toward the pedestrian road determines that the operating state of the unmanned vehicle is deflected away from the pedestrian road.
  • the vehicle controller executes a control command corresponding to the operating state according to the determined operating state, so that the unmanned vehicle can accurately adjust its operating state according to the control command.
  • the omnidirectional laser scanning of the obstacles is performed in a fixed position manner. Since the static laser radar does not need to perform three-dimensional rotary operation, only fixed-point detection is needed, so that the scanning omission is avoided, and the recognition accuracy of the obstacle is improved; the second aspect is that the static laser radar does not need to be dynamically rotated. The operation thus extends the life of the laser radar, provides higher stability, and reduces the burden on the unmanned vehicle, for example, reducing the heating energy consumption of the vehicle body. In the third aspect, compared with the rotary laser radar, the cost of the static laser radar is greatly reduced, so that the obstacle recognition accuracy is improved, and the manufacturing cost of the unmanned vehicle is also indirectly reduced.
  • FIG. 4 shows a specific implementation flow of an obstacle recognition method S103 for an unmanned vehicle provided by an embodiment of the present application, which is described in detail below;
  • S1031 Perform recognition processing on the second laser signal received in each direction to generate a three-dimensional environment image based on point cloud information.
  • the point cloud information refers to a set of vectors in a three-dimensional coordinate system. These vectors can represent the outer surface shape of the obstacle in the form of three-dimensional coordinates.
  • a rotatable camera located outside the driverless vehicle is activated.
  • the camera is configured to capture an environmental image of a specified orientation according to a control command received in real time.
  • the static lidar that collects the part of the point cloud information is determined to be based on the static lidar on the unmanned vehicle.
  • the preset position and angle control the camera to shoot in the direction emitted by the static laser radar to obtain a real-time captured image corresponding to the image area missing the point cloud information.
  • An obstacle is detected in the real-time captured image by a preset feature detection algorithm.
  • the feature detection algorithm is configured to detect whether there is a feature of the preset obstacle in the real-time captured image.
  • the method for detecting whether an obstacle exists in an image may be another image recognition algorithm, which is not limited herein.
  • the real-time image corresponding to the region where the point cloud information is missing in the three-dimensional environment image is captured by the camera, and the obstacle recognition processing is performed on the real-time captured real-time image, so that it is difficult to accurately identify the obstacle in the static lidar.
  • it can use the camera to realize the detection of the second line of defense, ensuring that even if the diffuse strip on the pedestrian road is missing, or if the reflected laser signal is not received, it can accurately take the image from the real-time. It is confirmed whether the obstacle exists or not, and the problem that the unmanned vehicle collides with the obstacle due to the untimely obstacle avoidance is avoided, thereby ensuring the safe driving of the unmanned vehicle and improving the safety performance.
  • the obstacle recognition method for an unmanned vehicle further includes: acquiring visibility information of a current time and a traveling speed of the unmanned vehicle; and according to the visibility information and the driving Speed, the sampling frequency of the static lidar is adjusted within a preset sampling frequency interval.
  • the real-time traveling speed of the unmanned vehicle is determined according to the output signal of the vehicle speed sensor of the unmanned vehicle.
  • the above visibility information specifically refers to a person with normal vision who can see the maximum distance of the target contour under current weather conditions.
  • weather processes such as rain, fog, haze, dust storms, etc.
  • the transparency of the atmosphere will decrease.
  • the visibility is poor, the penetration ability of the laser signal is greatly reduced, and the scanning omission of obstacles is prone to occur.
  • the visibility detected in the current environment can be detected by a visibility observer preset on an unmanned vehicle. Through the data transmission interface provided by the visibility observer, the visibility information transmitted by the visibility observer in real time can be received.
  • the third-party application server in the cloud may also be connected by way of wireless communication to receive real-time visibility information returned by the third-party application server.
  • the third-party server is used to store background data of the weather forecasting software and is used to provide an external calling service.
  • the obstacle recognition method provided by the embodiment of the present application can obtain the real-time visibility information through other existing methods, which is not limited herein.
  • each static laser radar has a preset sampling frequency interval, for example, the preset sampling frequency interval is 2000 Hz to 20,000 Hz.
  • the sampling frequency of the static lidar is adaptively increased within the sampling frequency interval, so that the sampling frequency of the static laser radar is larger than the original sampling frequency.
  • the sampling frequency of the static lidar is adaptively reduced within the sampling frequency interval, so that the sampling frequency of the static laser radar is smaller than the original sampling frequency.
  • the adjustment coefficient matching the interval level is determined according to the visibility level and the interval level at which the driving speed is respectively, and the original sampling frequency of the static lidar is adjusted according to the determined adjustment coefficient.
  • the adjustment coefficient is negative, the sampling frequency of the static lidar is lowered; when the adjustment coefficient is positive, the sampling frequency of the static lidar is lowered.
  • the larger the absolute value of the adjustment coefficient the larger the adjustment range of the sampling frequency.
  • the sampling frequency of the static lidar is adjusted according to the severity of the weather and the real-time traveling speed of the unmanned vehicle in a preset sampling frequency interval, so that the static laser radar can realize the sampling frequency.
  • Adaptability automatic adjustment under the condition of low performance requirements, by reducing the sampling frequency of static lidar, avoiding the problem of excessive energy consumption caused by long-term use of high sampling frequency; only in meeting performance requirements Under the condition, the sampling frequency of the laser signal is increased, and the point cloud information with higher precision can be obtained, thereby improving the detection efficiency and the accuracy of the obstacle. Therefore, the obstacle recognition method provided by the embodiment of the present application is better. A balance between reliability accuracy and heat consumption is achieved.
  • the above obstacle recognition method for an unmanned vehicle further includes:
  • S105 If receiving the vehicle stop command, controlling the static laser radar respectively located on the left and right sides of the door to emit the first laser signal, wherein a scanning area of the static lidar on the left and right sides of the door is the The drop-off area of the door.
  • FIG. 6 is a top view of a driverless bus with static laser radars disposed on both sides of the vehicle door according to another embodiment of the present application.
  • the driverless bus shown in FIG. 6 has a door 4, and at least one static laser radar 5 is disposed on each of the left and right sides of the door 4, and the direction of the static laser radar 5 is directed toward the getting-off region 6 of the door 4.
  • the passenger gets off the unmanned bus it is necessary to cross the door 4 and reach the getting off area 6, and the passenger who has not got on the bus can enter the driverless bus from the getting off area 6.
  • a static laser radar is disposed on each of the left and right sides of each door of the unmanned vehicle. If a vehicle stop command is received, the static laser radars on the left and right sides of the door are activated to emit a laser signal to the getting-off area of the door.
  • S106 If receiving the third laser signal obtained after the first laser signal is reflected, determining that there is an obstacle in the getting-off area of the door, and controlling the working state of the door to remain in a closed state.
  • the static lidar will receive the laser signal emitted by the obstacle. If the reflected laser signal is received within the preset time period, the working state of the door is kept closed; if the reflected laser signal is not received within the preset time period, it is determined that there is no obstacle in the current getting off area. The object triggers the door opening event to switch the working state of the door from the closed state to the open state.
  • the method further includes:
  • the infrared sensing device on the top side of the door 4 is activated to cause the infrared sensing device to detect the door region 7. If it is detected that there is an infrared signal in the door area 7, the door of the driverless vehicle is temporarily closed, and after a preset period of time, the execution is returned to the infrared sensing device to detect the door area; if it is detected that the door area 7 does not have an infrared signal, Then, the working state of the door is switched from the open state to the closed state.
  • the present application by statically installing the static laser radar at a position and an angle capable of covering the area where the vehicle is disembarked, it is ensured that when the vehicle stop command is detected, it is possible to automatically determine whether there is an obstacle in the getting-off area, but only in the obstacle.
  • the door of the driverless vehicle is activated to avoid the accidental risk of the obstacle to the passenger who gets off the vehicle; the door area is detected by controlling the infrared sensing device on the top side of the door, only when the infrared signal is not detected. The door is closed, and the passenger can be caught when the door is closed. Therefore, the obstacle recognition method provided by the embodiment of the present application can better protect the personal safety of the passenger.
  • FIG. 7 is a structural block diagram of an obstacle recognition apparatus for an unmanned vehicle provided by an embodiment of the present application, which is used for operating an unmanned vehicle according to the embodiment of FIGS. 1 to 6 of the present application. Obstacle identification method. For the convenience of explanation, only the parts related to the present embodiment are shown.
  • the apparatus includes:
  • the control unit 71 is configured to control the plurality of static laser radars on the unmanned vehicle to respectively transmit the first laser signals, and the plurality of static laser radars are respectively located around the body of the unmanned vehicle.
  • the receiving unit 72 is configured to receive the second laser signal obtained after the first laser signal is reflected.
  • the first obtaining unit 73 is configured to perform identification processing on the second laser signal received in each direction to acquire obstacle information.
  • the sending unit 74 is configured to send the obstacle information to the vehicle controller, so that the vehicle controller determines an operating state of the unmanned vehicle based on the obstacle information.
  • the first obtaining unit 73 includes:
  • control subunit configured to control a camera outside the unmanned vehicle to start shooting, if a location area where the missing point cloud information exists in the three-dimensional environment image is detected, to obtain a real-time captured image of the location area.
  • a detecting subunit configured to detect whether an obstacle exists in the real-time captured image.
  • control unit 71 is specifically configured to:
  • the convex mirror is located inside the static laser radar.
  • the device further includes:
  • the second obtaining unit 74 is configured to acquire visibility information of the current time and a traveling speed of the unmanned vehicle.
  • the adjusting unit 75 is configured to adjust a sampling frequency of the static lidar in a preset sampling frequency interval according to the visibility information and the traveling speed.
  • the device further includes:
  • a transmitting unit configured to: when the vehicle parking instruction is received, control the static laser radar respectively located on the left and right sides of the door to emit the first laser signal, wherein the scanning area of the static lidar on the left and right sides of the door It is the getting off area of the door.
  • a door closing unit configured to determine that there is an obstacle in the getting-off area of the door if the third laser signal obtained by the reflection of the first laser signal is received, and control the working state of the door to remain closed .
  • a door opening unit configured to: if the third laser signal obtained by the reflection of the first laser signal is not received, determine that there is no obstacle in the getting-off area of the door, and switch the working state of the door to Open state.
  • the omnidirectional laser scanning of the obstacles is performed in a fixed position manner. Since the static laser radar does not need to perform three-dimensional rotary operation, only fixed-point detection is needed, so that the scanning omission is avoided, and the recognition accuracy of the obstacle is improved; the second aspect is that the static laser radar does not need to be dynamically rotated. The operation thus extends the life of the laser radar, provides higher stability, and reduces the burden on the unmanned vehicle, for example, reducing the heating energy consumption of the vehicle body. In the third aspect, compared with the rotary laser radar, the cost of the static laser radar is greatly reduced, so that the obstacle recognition accuracy is improved, and the manufacturing cost of the unmanned vehicle is also indirectly reduced.
  • FIG. 8 is a schematic diagram of a terminal device according to an embodiment of the present application.
  • the terminal device 8 of this embodiment includes a processor 80, a memory 81, and a computer program 82, such as a voiceprint recognition program, stored in the memory 81 and operable on the processor 80.
  • the processor 80 executes the computer program 82 to implement the steps in the above-described embodiments of the obstacle recognition method for an unmanned vehicle, such as steps 101 to 104 shown in FIG.
  • the processor 80 when executing the computer program 82, implements the functions of the various modules/units in the various apparatus embodiments described above, such as the functions of the units 71-74 shown in FIG.
  • the computer program 82 can be partitioned into one or more modules/units that are stored in the memory 81 and executed by the processor 80 to complete This application.
  • the one or more modules/units may be a series of computer program instruction segments capable of performing a particular function, the instruction segments being used to describe the execution of the computer program 82 in the terminal device 8.
  • the terminal device 8 may be a computing device such as a desktop computer, a notebook, a palmtop computer, or a cloud server.
  • the terminal device 8 may include, but is not limited to, a processor 80 and a memory 81. It will be understood by those skilled in the art that FIG. 8 is merely an example of the terminal device 8, and does not constitute a limitation on the terminal device 8, and may include more or less components than those illustrated, or combine some components, or different components.
  • the terminal device 8 may further include an input/output device, a network access device, a bus, and the like.
  • the so-called processor 80 can be a central processing unit (Central Processing Unit, CPU), can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the memory 81 may be an internal storage unit of the terminal device 8, such as a hard disk or a memory of the terminal device 8.
  • the memory 81 may also be an external storage device of the terminal device 8, such as a plug-in hard disk provided on the terminal device 8, a smart memory card (SMC), and a secure digital (SD). Card, flash card, etc. Further, the memory 81 may also include both an internal storage unit of the terminal device 8 and an external storage device.
  • the memory 81 is used to store the computer program and other programs and data required by the terminal device 8.
  • the memory 81 can also be used to temporarily store data that has been output or is about to be output.
  • each functional module and module described above is exemplified. In practical applications, the above functions may be assigned to different functional modules according to needs.
  • the module is completed by dividing the internal structure of the device into different functional modules or modules to perform all or part of the functions described above.
  • Each functional module and module in the embodiment may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module, and the integrated module may be implemented by hardware.
  • Formal implementation can also be implemented in the form of software functional modules.
  • the specific names of the respective functional modules and modules are only for the purpose of facilitating mutual differentiation, and are not intended to limit the scope of protection of the present application.
  • modules and algorithm steps of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the solution. A person skilled in the art can use different methods to implement the described functions for each particular application, but such implementation should not be considered to be beyond the scope of the present application.
  • the disclosed apparatus and method may be implemented in other manners.
  • the system embodiment described above is merely illustrative.
  • the division of the module or module is only a logical function division.
  • there may be another division manner for example, multiple modules or components may be used. Combinations can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or module, and may be in electrical, mechanical or other form.
  • the modules described as separate components may or may not be physically separated.
  • the components displayed as modules may or may not be physical modules, that is, may be located in one place, or may be distributed to multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.
  • the above integrated modules can be implemented in the form of hardware or in the form of software functional modules.
  • the integrated modules if implemented in the form of software functional modules and sold or used as separate products, may be stored in a computer readable storage medium.
  • the medium includes a plurality of instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the methods described in the various embodiments of the embodiments of the present application.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), and a random access memory (RAM, Random Access).

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Electromagnetism (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)
  • Optical Radar Systems And Details Thereof (AREA)

Abstract

本方案适用于车辆技术领域，提供了一种用于无人驾驶车辆的障碍物识别方法、终端设备及介质，该方法包括：控制无人驾驶车辆上的多个静态激光雷达分别发射第一激光信号，多个所述静态激光雷达分别位于所述无人驾驶车辆的车身四周；接收所述第一激光信号经过反射后所得到的第二激光信号；对各个方向上接收得到的所述第二激光信号进行识别处理，以获取障碍物信息；将所述障碍物信息发送至车辆控制器，以使车辆控制器基于所述障碍物信息，确定出所述无人驾驶车辆的运行状态。本方案实现了以位置固定的方式来进行障碍物的全方位激光扫描，避免了出现扫描遗漏的情况，提高了障碍物的识别准确率，并且，延长了激光雷达的使用寿命，使其具有更高的稳定性。

Description

用于无人驾驶车辆的障碍物识别方法、装置及终端设备 技术领域
本申请属于车辆技术领域，尤其涉及一种用于无人驾驶车辆的障碍物识别方法、终端设备及介质。
 背景技术
无人驾驶汽车是通过车载传感系统感知道路环境，自动规划行车路线并控制车辆到达预定目标的智能汽车。由于其能够感知道路信息、车辆位置和障碍物信息，控制车辆的转向和速度，因此才保证了无人驾驶汽车能够安全、可靠地在道路上行驶。
目前，市面上的无人驾驶汽车一般通过实时三维旋转式的激光雷达来对障碍物进行检测。此类激光雷达以动态扫描的方式对环境进行感知。具体地，激光雷达内部的发射/接收组件和旋转镜面结合在一起，通过旋转镜面，能够实现90到180度的视角。镜面不仅能够反射二极管发出去的光，而且也能把反射回来的光再反射给接收器，从而使得安装在二极管附近的光子探测器能够探测出返回来的信号，以收集得到大量的点云，进而生成周围环境的三维图像。然而，虽然上述激光雷达可在一个维度上高度旋转，更新频率可以达到10Hz，但在垂直方向上，其视场角依然很小，大概为20度左右，因此，即使在水平维度上进行高速旋转，做到每秒扫描10次，发射2000个激光点，但最终得到的点云中，垂直方向的点分辨率也是较为稀疏的。由此，基于点云来识别障碍物时，仍然可能存在一些扫描盲区，例如，对一些较为细小的远处电线杆或者近处围栏等障碍物造成扫描遗漏，从而降低了障碍物检测的准确率，使得在实际环境中出现障碍物时，无人驾驶车辆依然判定当前无障碍物出现而持续前进的情况出现，因此，降低了无人驾驶车辆的安全性能。
 技术问题
有鉴于此，本申请实施例提供了一种用于无人驾驶车辆的障碍物识别方法、装置、终端设备及介质，以解决目前无人驾驶车辆的障碍物检测准确率以及安全性能均较为低下的问题。
 技术解决方案
本申请实施例的第一方面，提供了一种用于无人驾驶车辆的障碍物识别方法，包括：
控制无人驾驶车辆上的多个静态激光雷达分别发射第一激光信号，多个所述静态激光雷达分别位于所述无人驾驶车辆的车身四周；
接收所述第一激光信号经过反射后所得到的第二激光信号；
对各个方向上接收得到的所述第二激光信号进行识别处理，以获取障碍物信息；
将所述障碍物信息发送至车辆控制器，以使所述车辆控制器基于所述障碍物信息，确定出所述无人驾驶车辆的运行状态。
本申请实施例的第二方面，提供了一种用于无人驾驶车辆的障碍物识别装置，包括：
控制单元，用于控制无人驾驶车辆上的多个静态激光雷达分别发射第一激光信号，多个所述静态激光雷达分别位于所述无人驾驶车辆的车身四周；
接收单元，用于接收所述第一激光信号经过反射后所得到的第二激光信号；
获取单元，用于对各个方向上接收得到的所述第二激光信号进行识别处理，以获取障碍物信息；
发送单元，用于将所述障碍物信息发送至车辆控制器，以使所述车辆控制器基于所述障碍物信息，确定出所述无人驾驶车辆的运行状态。
本申请实施例的第三方面，提供了一种终端设备，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时，实现如第一方面所述的用于无人驾驶车辆的障碍物识别方法的步骤。
本申请实施例的第四方面，提供了一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被至少一个处理器执行时，实现如第一方面所述的用于无人驾驶车辆的障碍物识别方法的步骤。
 有益效果
本申请实施例中，通过控制位于无人驾驶车辆的车身四周上的多个静态激光雷达分别发射激光信号，实现了以位置固定的方式来进行障碍物的全方位激光扫描。由于静态激光雷达不需要执行三维旋转式操作，而只需定点检测，因此避免了出现扫描遗漏的情况，提高了障碍物的识别准确率；第二方面，正因为静态激光雷达不需要进行动态旋转式操作，因而减少了对无人驾驶车辆的负担，例如，降低了车身的发热能耗等，同时也延长了激光雷达的使用寿命，使之具有更高的稳定性。第三方面，相对于旋转式激光雷达来说，静态激光雷达的成本大大降低，故在提高了障碍物识别准确率的同时，也间接降低了无人驾驶车辆的制造成本。
 附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的用于无人驾驶车辆的障碍物识别方法的实现流程图；
图2本申请实施例提供的车身四周分别设置有静态激光雷达的无人驾驶公交的俯视图；
图3是本申请实施例提供的静态激光雷达的内部结构图；
图4是本申请实施例提供的用于无人驾驶车辆的障碍物识别方法S103的具体实现流程图；
图5是本申请又一实施例提供的用于无人驾驶车辆的障碍物识别方法的实现流程图；
图6是本申请又一实施例提供的车门两侧分别设置有静态激光雷达的无人驾驶公交的俯视图；
图7是本申请实施例提供的用于无人驾驶车辆的障碍物识别装置的结构框图；
图8是本申请实施例提供的终端设备的示意图。
 本发明的实施方式
以下描述中，为了说明而不是为了限定，提出了诸如特定系统结构、技术之类的具体细节，以便透彻理解本申请实施例。然而，本领域的技术人员应当清楚，在没有这些具体细节的其它实施例中也可以实现本申请。在其它情况中，省略对众所周知的系统、装置、电路以及方法的详细说明，以免不必要的细节妨碍本申请的描述。
图1示出了本申请实施例提供的用于无人驾驶车辆的障碍物识别方法的实现流程，该方法流程包括步骤S101至S104。各步骤的具体实现原理如下。
S101：控制无人驾驶车辆上的多个静态激光雷达分别发射第一激光信号，多个所述静态激光雷达分别位于所述无人驾驶车辆的车身四周。
本申请实施例中，静态激光雷达为非旋转式激光雷达，例如可以是微机电系统（MicroElectroMechanicalSystem，MEMS）微动的激光雷达等。静态激光雷达位于无人驾驶车辆的车身四周，使得无人驾驶车辆的每一角部都分别存在至少一个静态激光雷达。
优选地，上述无人驾驶车辆包括但不限于无人驾驶公交以及无人驾驶轿车。
示例性地，图2为本申请实施例提供的车身四周分别设置有静态激光雷达的无人驾驶公交的俯视图。如图2所示，该无人驾驶公交的车头1和车尾2的左右两侧都分别设置有一静态激光雷达3，静态激光雷达3可以位于车身的四个角部，也可以位于车身各个侧面的交线处。本示例中，控制静态激光雷达所发出的激光信号与车身长度延长线的夹角为固定值α
本申请实施例中，当接收到无人驾驶车辆的运行指令时，启动上述各个静态激光雷达，以控制每一静态激光雷达内部的激光发射器发出激光信号。由于各个静态激光雷达所处的位置角度不同，故对于整个无人驾驶车辆而言，可发出多种方向的激光信号，使得激光信号的扫描位置能够覆盖可能影响无人驾驶车辆行驶的各个斜方区域。
优选地，本申请实施例中，在接收到激光信号发射指令时，控制无人驾驶车辆上每一静态激光雷达的激光光源经凸面反射镜反射之后形成激光信号，由此使得激光信号经过镜面反射后才投射至外界的目标物体。其中，所述凸面镜位于所述静态激光雷达内部。
请一并参照图3，图3为本申请实施例提供的静态激光雷达的内部结构示意图。图3中，静态激光雷达内部的激光发射器31作为激光光源，发出携带上述激光信号的激光束。对于凸面反射镜32来说，其凸面一侧所朝的方向与激光束的发射方向相对。基于激光沿直线传播的特性，激光发射器31以较小的散射角度发出激光束后，激光束所包含的各道激光线会投射于凸面反射镜32上的不同位置点，并在各个位置点发射镜面反射后，投射于静态激光雷达外部的目标物体。
本申请实施例中，通过在静态激光雷达内部设置一与激光束发射方向相对的凸面镜，使得激光光源所发出的激光信号能够经过凸面反射镜的弧度面进行发射后再投射至外界环境，相对于静态激光雷达所原有的将激光信号直接投射至外界环境的方式来说，扩大了静态激光雷达的扫描区域，使得固定位置的静态激光雷达所发出的激光信号能够覆盖范围更广的目标物体，进而达到更为全面的障碍物检测效果，提高了障碍物检测效率。
S102：接收所述第一激光信号经过反射后所得到的第二激光信号。
静态激光雷达内部的激光发射器作为激光光源，发射出的脉冲激光信号将打到该静态激光雷达的扫描区域。若扫描区域中存在障碍物，例如，存在树木、道路、桥梁或建筑物时，则会引起脉冲激光信号发生散射现象，此时，一部分激光信号会反射至静态激光雷达内部的接收器上。
对于每次发射的激光信号，通过静态激光雷达内置的接收器，接收其经过反射后所回波的激光信号。
作为本申请实施例的一个优选实施方式，由于无人驾驶车辆需要正确识别其行驶道路，以避免无人驾驶车辆行驶至人行道路上。因此，在人行道的侧面，贴上漫反光条，则上述S102具体包括：
接收所述第一激光信号经过漫反光条反射后所得到的第二激光信号，其中，所述漫反光条覆盖于人行道路的侧面。
上述人行道路，是指位于车辆行驶道路两侧且高于车辆行驶道路的其他道路；人行道路的侧面即人行道路的高度线所在的平面。将上述静态激光雷达预设于无人驾驶车辆的车身四周，并对车头左右两侧的静态激光雷达的高度及其激光信号的发射角度进行调整，以使得车头左右两侧的每一静态激光雷达能够清楚完整地扫描到人行道路上的漫反光条，保证在运行过程中，无论无人驾驶车辆的处于汽车行驶道路的哪一位置，其车头左右两侧发出的激光信号都能完全覆盖漫反光条的最小角度。另一方面，对车尾左右两侧的静态激光雷达的高度及其激光信号的发射角度进行调整，使得车尾左右两侧的每一静态激光雷达能够具有更高的概率能够地扫描到无人驾驶车辆后方的其他车辆，进而可以在未检测到后方车辆的情况下，进行安全的变道操作。
本申请实施例中，漫反光条的表面覆盖有毛面材料，毛面材料用于漫反射905纳米的激光信号。具体地，漫反光条所采用的毛面材料由二氧化钛R-930、硫酸钡以及玻璃微珠混合而成。在混合过程中，硫酸钡作为分散剂，玻璃微珠作为填充物。其中，玻璃微珠是指，以玻璃砂为原料，并对该原料进行高温融熔处理后所得到一种玻璃圆珠。玻璃微珠的直径在75微米至1400微米之间，折射率为1.50至1.64之间。
当激光信号在漫反光条中发生散射时，散射光波长和粒径满足以下关系：
λ=d/k；k=0.9(m 2+2)/nR·π·（m 2-1）（1）
式子（1），nR为树脂折光指数，m为毛面材料的散射能力，k为与m和nR关联的预设常量，d为毛面材料的粒径，λ为散射的激光信号的波长。基于公式（1）可计算得出，上述用于发射905纳米激光信号的毛面材料中，二氧化钛的粒径在362纳米至394纳米之间。
本申请实施例中，由于硫酸钡仅作为分散剂使用，因而可以使得二氧化钛在混合过程中能够得到均匀分散，提高了漫反光条的漫反射能力，降低了毛面材料的成本。通过在毛面材料的制作过程中加入玻璃微珠，一定程度上也提高漫反光条的漫反射能力以及逆反射能力。由于人行道路的侧面设置有漫反光条，因而基于漫反光条的漫反射作用，使得静态激光雷达发出的激光信号经过漫反光条反射后，能够获得更准确、显著的障碍物识别效果，由此也提高了障碍物的识别准确率，进而提高了无人驾驶车辆的安全性能。
S103：对各个方向上接收得到的所述第二激光信号进行识别处理，以获取障碍物信息。
对于接收到的从障碍物反射回来的激光信号，根据其激光反射强度以及其三维坐标数据，获取该激光信号所对应得到的点云信息。由于点云是在同一空间坐标系下表达障碍物空间分布和障碍物表面特性的海量点云数据，因而通过多个位置的静态激光雷达所接收得到的回波的激光信号，可获得障碍物的状态信息。上述状态信息包括但不限于障碍物相对于无人驾驶车辆的距离、方位、高度、速度、姿态以及形状等，例如，人行道路相对于所述无人驾驶车辆的方位及距离。其中，障碍物的距离信息通过激光测距原理计算得出。
S104：将所述障碍物信息发送至车辆控制器，以使所述车辆控制器基于所述障碍物信息，确定出所述无人驾驶车辆的运行状态。
本申请实施例中，将实时更新的障碍物信息发送至车辆控制器。车辆控制器用于确定无人驾驶车辆的运行状态，运行状态包括行驶速度、各车轮的转向信息以及急刹信息等。
示例性地，若车辆控制器所获取得到的障碍物信息为，前方1米处存在障碍物，则根据预设的避障算法，可确定出无人驾驶车辆的运行状态为急刹车；若障碍物信息具体为，车身所在直线的延伸方向朝向人行道路的方向，则确定无人驾驶车辆的运行状态为向远离人行道路的方向偏转。
车辆控制器根据确定出的运行状态，执行与运行状态对应的控制指令，以使得无人驾驶车辆能够根据控制指令准确调节自身的运行状态。
本申请实施例中，通过控制位于无人驾驶车辆的车身角部上的多个静态激光雷达分别发射第一激光信号，实现了以位置固定的方式来进行障碍物的全方位激光扫描。由于静态激光雷达不需要执行三维旋转式操作，而只需定点检测，因此避免出现了扫描遗漏的情况，提高了障碍物的识别准确率；第二方面，正因为静态激光雷达不需要进行动态旋转式操作，因而延长了激光雷达的使用寿命，具有更高的稳定性，且减少了对无人驾驶车辆的负担，例如，降低了车身的发热能耗等。第三方面，相对于旋转式激光雷达来说，静态激光雷达的成本大大降低，故在提高了障碍物识别准确率的同时，也间接降低了无人驾驶车辆的制造成本。
作为本申请的一个实施例，图4示出了本申请实施例提供的用于无人驾驶车辆的障碍物识别方法S103的具体实现流程，详述如下；
S1031：对各个方向上接收得到的所述第二激光信号进行识别处理，以生成基于点云信息的三维环境图像。
本申请实施例中，点云信息是指在三维坐标系统中的一组向量的集合。这些向量能够以三维坐标的形式来表示障碍物的外表面形状。
通过控制静态激光雷达持续发出激光信号，能够不断扫车身外侧的障碍物，从而得到障碍物上各个目标点所反射的激光信号，将此激光信号所对应的点云信息进行成像处理后，可得到精确的三维环境图像。
S1032：若检测到所述三维环境图像中存在缺失点云信息的位置区域，则控制所述无人驾驶车辆外侧的摄像头启动拍摄，得到该位置区域的实时拍摄图像。
由于静态激光雷达所发射的激光信号可能存在穿透力较弱的问题，因此，若因穿透力不足或者因其他因素原因而导致无法扫描得到某一位置点的点云信息时，则在上述三维环境图像中，将会出现部分区域缺失的情况。此时，启动位于无人驾驶车辆外侧的可旋转式的摄像头。上述摄像头用于根据实时接收到的控制指令，拍摄指定方位的环境图像。
本申请实施例中，根据三维环境图像中缺失点云信息的区域的附近其他点云信息，确定出采集这部分点云信息的静态激光雷达，以根据该静态激光雷达在无人驾驶车辆上的预设位置及角度，控制摄像头向该静态激光雷达所发射的方向进行拍摄，以得到缺失点云信息的图像区域所对应实时拍摄图像。
S1033：检测所述实时拍摄图像中是否存在障碍物。
通过预设的特征检测算法，检测实时拍摄图像中是否存在障碍物。特征检测算法用于检测实时拍摄图像中是否存在预设障碍物的特征。其中，检测图像中是否存在障碍物的方法还可以是其他图像识别算法，在此不作限定。
本申请实施例中，通过控制摄像头拍摄三维环境图像中缺失点云信息的区域所对应的实景图像，并对该实时拍摄的实景图像进行障碍物识别处理，使得在静态激光雷达难以准确识别障碍物是否存在时，能够利用摄像头来实现第二道防线的检测，保证了即使人行道路上的漫反光条缺失，或者，在没有接收到反射回来的激光信号时，也能准确地从实时拍摄图像中进一步地确认障碍物是否存在，避免了因避障不及时而导致无人驾驶车辆与障碍物发生碰撞的问题，因此，保障了无人驾驶车辆的安全行驶，提高了安全性能。
作为本申请的另一实施例，上述用于无人驾驶车辆的障碍物识别方法还包括：获取当前时刻的能见度信息以及所述无人驾驶车辆的行驶速度；根据所述能见度信息以及所述行驶速度，在预设的采样频率区间内，调节所述静态激光雷达的采样频率。
本申请实施例中，根据无人驾驶车辆的车速传感器的输出信号，确定出无人驾驶车辆的实时行驶速度。
上述能见度信息，具体是指，具有正常视力的人在当前天气条件下所能够看清楚目标轮廓的最大距离。当出现降雨、雾、霾、沙尘暴等天气过程时，大气透明度将会降低，此时能见度较差，激光信号的穿透能力大大降低，因而容易出现障碍物的扫描遗漏。
作为本申请的一个具体实施示例，可通过无人驾驶车辆上所预先设置的能见度观测仪，检测当前环境下所检测得到的能见度。通过能见度观测仪所提供的数据传输接口，可接收能见度观测仪所实时传输的能见度信息。
作为本申请的另一个具体实施示例，还可通过无线通信的方式，与云端的第三方应用服务器进行连接，以接收第三方应用服务器所返回的实时能见度信息。其中，上述第三方服务器用于存储天气预报软件的后台数据，并用于提供外部调用服务。
值得注意的是，除了上述能见度信息的获取方式之外，本申请实施例所提供的障碍物识别方法还可以通过其他现有的方式来获取实时能见度信息，在此不作限定。
本申请实施例中，每一静态激光雷达具有一预设的采样频率区间，例如，预设的采样频率区间为2000赫兹至20000赫兹。当能见度越低或行驶速度越高时，在采样频率区间之内，适应性地增大静态激光雷达的采样频率，以使静态激光雷达的采样频率大于原有采样频率。当能见度越高或行驶速度越低时，在采样频率区间之内，适应性地降低静态激光雷达的采样频率，以使静态激光雷达的采样频率小于原有采样频率。
优选地，根据能见度信息以及行驶速度分别所处的区间等级，确定与区间等级匹配的调节系数，并根据确定出的调节系数，对静态激光雷达的原有采样频率进行调整。当调节系数为负值时，降低静态激光雷达的采样频率；当调节系数为正值时，降低静态激光雷达的采样频率。并且，调节系数的绝对值越大，采样频率的调节幅度越大。
本申请实施例中，通过在预设的采样频率区间内，根据天气的恶劣程度以及无人驾驶车辆的实时行驶速度来对静态激光雷达的采样频率进行调整，使得静态激光雷达能够实现采样频率的适应性自动调整，在性能要求较低的情况之下，通过降低静态激光雷达的采样频率，避免了因长期采用高采样频率而带来的能耗过高的问题；而仅在满足性能需求的条件之下，才增大激光信号的采样频率，能够获取得到精度更高的点云信息，从而提高障碍物的检测效率以及准确率；因此，本申请实施例提供的障碍物识别方法较好地实现了可靠性精度以及能耗发热之间的平衡。
作为本申请的又一实施例，如图5所示，上述用于无人驾驶车辆的障碍物识别方法还包括：
S105：若接收到车辆停靠指令，则控制分别位于车门的左右两侧的所述静态激光雷达发射所述第一激光信号，其中，车门左右两侧的所述静态激光雷达的扫描区域为所述车门的下车区域。
请一并参照图6，图6为本申请又一实施例提供的车门两侧分别设置有静态激光雷达的无人驾驶公交的俯视图。如图6所示的无人驾驶公交具有一车门4，在车门4的左右两侧均分别设置有至少一个静态激光雷达5，静态激光雷达5的发射方向朝向车门4的下车区域6。乘客从无人驾驶公交下车时，需要跨越车门4并到达下车区域6，且未上车的乘客可以从下车区域6进入无人驾驶公交。
本申请实施例中，无人驾驶车辆的每一车门的左右两侧均设置有静态激光雷达。若接收到车辆停靠指令，则启动该处于车门的左右两侧的各个静态激光雷达，以向该车门的下车区域发出激光信号。
S106：若接收到所述第一激光信号经过反射后所得到的第三激光信号，则确定所述车门的下车区域存在障碍物，控制所述车门的工作状态保持为关闭状态。
S107：若未接收到所述第一激光信号经过反射后所得到的第三激光信号，则确定所述车门的下车区域不存在障碍物，将所述车门的工作状态切换为开启状态。
若车门的下车区域存在障碍物，则静态激光雷达会接收到经该障碍物所发射回来的激光信号。若在预设时长内，接收到反射回来的激光信号，则保持车门的工作状态为关闭状态；若在预设时长内，未接收到反射回来的激光信号，则确定当前下车区域不存在障碍物，故触发车门开启事件，以将车门的工作状态由关闭状态切换为开启状态。
优选地，如图6所示，本申请实施例中，在车门开启事件被触发之后，还包括：
若检测到车门关闭事件被触发，则启动车门4顶侧的红外感应装置，以使红外感应装置对车门区域7进行检测。若检测到车门区域7存在红外信号，则暂停关闭无人驾驶车辆的车门，并在预设时长后，返回执行令红外感应装置对车门区域进行检测；若检测到车门区域7不存在红外信号，则将车门的工作状态由开启状态切换为关闭状态。
本申请实施例中，通过在能够覆盖车门下车区域的位置和角度固定安装静态激光雷达，保证了在检测到车辆停靠指令时，能够自动判断下车区域是否存在障碍物，而仅在障碍物不存在时，才启动无人驾驶车辆的车门，避免了障碍物对下车乘客带来的意外风险；通过控制车门顶侧的红外感应装置对车门区域进行检测，仅在未检测到红外信号时，才关闭车门，能够避免车门关闭时会夹到乘客，因此，本申请实施例提供的障碍物识别方法较好地保障了乘客的人身安全。
应理解，上述实施例中各步骤的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
图7示出了本申请实施例提供的用于无人驾驶车辆的障碍物识别装置的结构框图，该装置用于运行本申请图1至图6实施例所述的用于无人驾驶车辆的障碍物识别方法。为了便于说明，仅示出了与本实施例相关的部分。
参照图7，该装置包括：
控制单元71，用于控制无人驾驶车辆上的多个静态激光雷达分别发射第一激光信号，多个所述静态激光雷达分别位于所述无人驾驶车辆的车身四周。
接收单元72，用于接收所述第一激光信号经过反射后所得到的第二激光信号。
第一获取单元73，用于对各个方向上接收得到的所述第二激光信号进行识别处理，以获取障碍物信息。
发送单元74，用于将所述障碍物信息发送至车辆控制器，以使所述车辆控制器基于所述障碍物信息，确定出所述无人驾驶车辆的运行状态。
可选地，所述第一获取单元73包括：
生成子单元，用于对各个方向上接收得到的的所述第二激光信号进行识别处理，以生成基于点云信息的三维环境图像。
控制子单元，用于若检测到所述三维环境图像中存在缺失点云信息的位置区域，则控制所述无人驾驶车辆外侧的摄像头启动拍摄，得到该位置区域的实时拍摄图像。
检测子单元，用于检测所述实时拍摄图像中是否存在障碍物。
可选地，所述控制单元71具体用于：
控制无人驾驶车辆上的多个静态激光雷达的激光光源经凸面反射镜反射之后形成所述第一激光信号；
其中，所述凸面镜位于所述静态激光雷达内部。
可选地，该装置还包括：
第二获取单元74，用于获取当前时刻的能见度信息以及所述无人驾驶车辆的行驶速度。
调节单元75，用于根据所述能见度信息以及所述行驶速度，在预设的采样频率区间内，调节所述静态激光雷达的采样频率。
可选地，该装置还包括：
发射单元，用于若接收到车辆停靠指令，则控制分别位于车门的左右两侧的所述静态激光雷达发射所述第一激光信号，其中，车门左右两侧的所述静态激光雷达的扫描区域为所述车门的下车区域。
车门关闭单元，用于若接收到所述第一激光信号经过反射后所得到的第三激光信号，则确定所述车门的下车区域存在障碍物，控制所述车门的工作状态保持为关闭状态。
车门开启单元，用于若未接收到所述第一激光信号经过反射后所得到的第三激光信号，则确定所述车门的下车区域不存在障碍物，将所述车门的工作状态切换为开启状态。
本申请实施例中，通过控制位于无人驾驶车辆的车身角部上的多个静态激光雷达分别发射第一激光信号，实现了以位置固定的方式来进行障碍物的全方位激光扫描。由于静态激光雷达不需要执行三维旋转式操作，而只需定点检测，因此避免出现了扫描遗漏的情况，提高了障碍物的识别准确率；第二方面，正因为静态激光雷达不需要进行动态旋转式操作，因而延长了激光雷达的使用寿命，具有更高的稳定性，且减少了对无人驾驶车辆的负担，例如，降低了车身的发热能耗等。第三方面，相对于旋转式激光雷达来说，静态激光雷达的成本大大降低，故在提高了障碍物识别准确率的同时，也间接降低了无人驾驶车辆的制造成本。
图8是本申请实施例提供的终端设备的示意图。如图8所示，该实施例的终端设备8包括：处理器80、存储器81以及存储在所述存储器81中并可在所述处理器80上运行的计算机程序82，例如声纹识别程序。所述处理器80执行所述计算机程序82时实现上述各个用于无人驾驶车辆的障碍物识别方法实施例中的步骤，例如图1所示的步骤101至104。或者，所述处理器80执行所述计算机程序82时实现上述各装置实施例中各模块/单元的功能，例如图7所示单元71至74的功能。
示例性的，所述计算机程序82可以被分割成一个或多个模块/单元，所述一个或者多个模块/单元被存储在所述存储器81中，并由所述处理器80执行，以完成本申请。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述所述计算机程序82在所述终端设备8中的执行过程。
所述终端设备8可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述终端设备8可包括，但不仅限于，处理器80、存储器81。本领域技术人员可以理解，图8仅仅是终端设备8的示例，并不构成对终端设备8的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如所述终端设备8还可以包括输入输出设备、网络接入设备、总线等。
所称处理器80可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器 (Digital Signal Processor，DSP)、专用集成电路 (Application Specific Integrated Circuit，ASIC)、现成可编程门阵列 (Field-Programmable Gate Array，FPGA) 或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
所述存储器81可以是所述终端设备8的内部存储单元，例如终端设备8的硬盘或内存。所述存储器81也可以是所述终端设备8的外部存储设备，例如所述终端设备8上配备的插接式硬盘，智能存储卡（Smart Media Card，SMC），安全数字（Secure Digital，SD）卡，闪存卡（Flash Card）等。进一步地，所述存储器81还可以既包括所述终端设备8的内部存储单元也包括外部存储设备。所述存储器81用于存储所述计算机程序以及所述终端设备8所需的其他程序和数据。所述存储器81还可以用于暂时地存储已经输出或者将要输出的数据。
所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，仅以上述各功能模块、模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块、模块完成，即将所述装置的内部结构划分成不同的功能模块或模块，以完成以上描述的全部或者部分功能。实施例中的各功能模块、模块可以集成在一个处理模块中，也可以是各个模块单独物理存在，也可以两个或两个以上模块集成在一个模块中，上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。另外，各功能模块、模块的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。上述系统中模块、模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的模块及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
在本申请所提供的实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的系统实施例仅仅是示意性的，例如，所述模块或模块的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个模块或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口，装置或模块的间接耦合或通讯连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的模块可以是或者也可以不是物理上分开的，作为模块显示的部件可以是或者也可以不是物理模块，即可以位于一个地方，或者也可以分布到多个网络模块上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能模块可以集成在一个处理模块中，也可以是各个模块单独物理存在，也可以两个或两个以上模块集成在一个模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。
所述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请实施例的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备（可以是个人计算机，服务器，或者网络设备等）或处理器（processor）执行本申请实施例各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器（ROM，Read-Only Memory）、随机存取存储器（RAM，Random Access Memory）、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围，均应包含在本申请的保护范围之内。

Claims (10)

  1. 一种用于无人驾驶车辆的障碍物识别方法，其特征在于，包括：
    控制无人驾驶车辆上的多个静态激光雷达分别发射第一激光信号，多个所述静态激光雷达分别位于所述无人驾驶车辆的车身四周；
    接收所述第一激光信号经过反射后所得到的第二激光信号；
    对各个方向上接收得到的所述第二激光信号进行识别处理，以获取障碍物信息；
    将所述障碍物信息发送至车辆控制器，以使所述车辆控制器基于所述障碍物信息，确定出所述无人驾驶车辆的运行状态。
  2. 如权利要求1所述的障碍物识别方法，其特征在于，所述对各个方向上接收得到的所述第二激光信号进行识别处理，以获取障碍物信息，包括：
    对各个方向上接收得到的的所述第二激光信号进行识别处理，以生成基于点云信息的三维环境图像；
    若检测到所述三维环境图像中存在缺失点云信息的位置区域，则控制所述无人驾驶车辆外侧的摄像头启动拍摄，得到该位置区域的实时拍摄图像；
    检测所述实时拍摄图像中是否存在障碍物。
  3. 如权利要求1或2所述的障碍物识别方法，其特征在于，所述控制无人驾驶车辆上的多个静态激光雷达分别发射第一激光信号，包括：
    控制无人驾驶车辆上的多个静态激光雷达的激光光源经凸面反射镜反射之后形成所述第一激光信号；
    其中，所述凸面镜位于所述静态激光雷达内部。
  4. 如权利要求1-3任一项所述的障碍物识别方法，其特征在于，还包括：
    获取当前时刻的能见度信息以及所述无人驾驶车辆的行驶速度；
    根据所述能见度信息以及所述行驶速度，在预设的采样频率区间内，调节所述静态激光雷达的采样频率。
  5. 如权利要求1-3任一项所述的障碍物识别方法，其特征在于，还包括：
    若接收到车辆停靠指令，则控制分别位于车门的左右两侧的所述静态激光雷达发射所述第一激光信号，其中，车门左右两侧的所述静态激光雷达的扫描区域为所述车门的下车区域；
    若接收到所述第一激光信号经过反射后所得到的第三激光信号，则确定所述车门的下车区域存在障碍物，控制所述车门的工作状态保持为关闭状态；
    若未接收到所述第一激光信号经过反射后所得到的第三激光信号，则确定所述车门的下车区域不存在障碍物，将所述车门的工作状态切换为开启状态。
  6. 一种用于无人驾驶车辆的障碍物识别装置，其特征在于，包括：
    控制单元，用于控制无人驾驶车辆上的多个静态激光雷达分别发射第一激光信号，多个所述静态激光雷达分别位于所述无人驾驶车辆的车身四周；
    接收单元，用于接收所述第一激光信号经过反射后所得到的第二激光信号；
    获取单元，用于对各个方向上接收得到的所述第二激光信号进行识别处理，以获取障碍物信息；
    发送单元，用于将所述障碍物信息发送至车辆控制器，以使所述车辆控制器基于所述障碍物信息，确定出所述无人驾驶车辆的运行状态。
  7. 如权利要求6所述的障碍物识别装置，其特征在于，所述获取单元包括：
    生成子单元，用于对各个方向上接收得到的的所述第二激光信号进行识别处理，以生成基于点云信息的三维环境图像；
    控制子单元，用于若检测到所述三维环境图像中存在缺失点云信息的位置区域，则控制所述无人驾驶车辆外侧的摄像头启动拍摄，得到该位置区域的实时拍摄图像；
    检测子单元，用于检测所述实时拍摄图像中是否存在障碍物。
  8. 如权利要求6或7所述的障碍物识别装置，其特征在于，所述控制单元具体用于：
    控制无人驾驶车辆上的多个静态激光雷达的激光光源经凸面反射镜反射之后形成所述第一激光信号；
    其中，所述凸面镜位于所述静态激光雷达内部。
  9. 一种终端设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，其特征在于，所述处理器执行所述计算机程序时实现如权利要求1至5任一项所述方法的步骤。
  10. 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1至5任一项所述方法的步骤。
PCT/CN2017/117730 2017-12-19 2017-12-21 用于无人驾驶车辆的障碍物识别方法、装置及终端设备 WO2019119350A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711375585.5A CN108318895B (zh) 2017-12-19 2017-12-19 用于无人驾驶车辆的障碍物识别方法、装置及终端设备
CN201711375585.5 2017-12-19

Publications (1)

Publication Number Publication Date
WO2019119350A1 true WO2019119350A1 (zh) 2019-06-27

Family

ID=62891795

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/117730 WO2019119350A1 (zh) 2017-12-19 2017-12-21 用于无人驾驶车辆的障碍物识别方法、装置及终端设备

Country Status (2)

Country Link
CN (1) CN108318895B (zh)
WO (1) WO2019119350A1 (zh)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111144211A (zh) * 2019-08-28 2020-05-12 华为技术有限公司 点云显示方法和装置
CN111443360A (zh) * 2020-04-20 2020-07-24 北京易控智驾科技有限公司 矿区无人驾驶系统道路边界自动采集装置和识别方法
CN112327308A (zh) * 2019-07-19 2021-02-05 阿里巴巴集团控股有限公司 物体检测方法、装置、系统及设备
CN112348777A (zh) * 2020-10-19 2021-02-09 深圳市优必选科技股份有限公司 人体目标的检测方法、装置及终端设备
CN112415538A (zh) * 2020-11-16 2021-02-26 重庆长安汽车股份有限公司 自动驾驶车辆绕行锥形筒的规划方法、系统及车辆
CN112445215A (zh) * 2019-08-29 2021-03-05 阿里巴巴集团控股有限公司 自动导引车行驶控制方法、装置及计算机系统
CN112485774A (zh) * 2020-11-26 2021-03-12 中国第一汽车股份有限公司 一种车载激光雷达标定方法、装置、设备及存储介质
CN112558103A (zh) * 2019-09-10 2021-03-26 奥迪股份公司 障碍物检测设备、障碍物的检测方法及装置、计算机设备
CN113095266A (zh) * 2021-04-19 2021-07-09 北京经纬恒润科技股份有限公司 一种角度识别方法、装置及设备
CN113467450A (zh) * 2021-07-01 2021-10-01 北京小狗吸尘器集团股份有限公司 无人机器控制方法、装置、计算机设备及存储介质
CN113985389A (zh) * 2021-09-30 2022-01-28 苏州浪潮智能科技有限公司 一种时间同步校准装置、自动识别路径设备、方法及介质
CN115327571A (zh) * 2022-07-29 2022-11-11 武汉理工大学 一种基于平面激光雷达的三维环境障碍物检测系统及方法
EP4091424A1 (en) * 2021-05-19 2022-11-23 AGCO International GmbH Residue spread monitoring
CN116412814A (zh) * 2023-06-12 2023-07-11 旷智中科(北京)技术有限公司 一种基于激光雷达的图像构建导航辅助系统

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109212540B (zh) * 2018-09-12 2024-07-26 阿波罗智能技术(北京)有限公司 基于激光雷达系统的测距方法、装置及可读存储介质
CN109506949B (zh) * 2018-11-12 2020-08-04 百度在线网络技术（北京）有限公司 无人驾驶车辆的物体识别方法、装置、设备及存储介质
CN109614886B (zh) * 2018-11-23 2022-06-24 中国人民解放军63966部队 一种无人/自主驾驶系统的性能评测方法及装置
CN110045382A (zh) * 2018-12-03 2019-07-23 阿里巴巴集团控股有限公司 车辆损伤检测的处理方法、装置、设备、服务器和系统
CN109633688B (zh) * 2018-12-14 2019-12-24 北京百度网讯科技有限公司 一种激光雷达障碍物识别方法和装置
CN109725328B (zh) * 2018-12-17 2023-04-28 云南昆船智能装备有限公司 一种基于激光雷达传感器的agv障碍物检测系统及方法
DK180562B1 (en) 2019-01-31 2021-06-28 Motional Ad Llc Merging data from multiple lidar devices
DE102019211739A1 (de) * 2019-08-06 2021-02-11 Ibeo Automotive Systems GmbH Lidar-Messsystem mit zwei Lidar-Messvorrichtungen
CN110522616A (zh) * 2019-09-02 2019-12-03 湖南工程学院 一种可穿戴智能导盲衣
CN110542908B (zh) * 2019-09-09 2023-04-25 深圳市海梁科技有限公司 应用于智能驾驶车辆上的激光雷达动态物体感知方法
CN110794831B (zh) * 2019-10-16 2023-07-28 深圳乐动机器人股份有限公司 一种控制机器人工作的方法及机器人
CN110803104A (zh) * 2019-11-07 2020-02-18 浙江鸿泉电子科技有限公司 车门盲区监测系统、监测方法及车辆
CN110979321B (zh) * 2019-12-30 2021-03-19 北京深测科技有限公司 一种用于无人驾驶车辆的障碍物躲避方法
CN111563975B (zh) * 2020-03-14 2021-12-07 青岛民航凯亚系统集成有限公司 区块链飞行器安全防护系统及方法
US11977165B2 (en) * 2020-08-10 2024-05-07 Waymo Llc Self-reflection filtering
CN112991511B (zh) * 2020-10-13 2023-03-24 中国汽车技术研究中心有限公司 一种点云数据的展示方法
CN112306061B (zh) * 2020-10-28 2024-06-21 深圳优地科技有限公司 一种机器人控制方法以及机器人
CN112666569B (zh) * 2020-12-01 2023-03-24 天津优控智行科技有限公司 一种无人驾驶系统激光雷达连续点云的压缩方法
CN112733923A (zh) * 2021-01-04 2021-04-30 上海高仙自动化科技发展有限公司 一种确定禁行区域的系统及机器人
CN112785704B (zh) * 2021-01-13 2024-07-26 北京小马慧行科技有限公司 语义地图的构建方法、计算机可读存储介质和处理器

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102779280A (zh) * 2012-06-19 2012-11-14 武汉大学 一种基于激光传感器的交通信息提取方法
CN102837658A (zh) * 2012-08-27 2012-12-26 北京工业大学 一种智能车辆中多激光雷达数据融合系统和方法
CN104899855A (zh) * 2014-03-06 2015-09-09 株式会社日立制作所 三维障碍物检测方法和装置
US20160018524A1 (en) * 2012-03-15 2016-01-21 GM Global Technology Operations LLC SYSTEM AND METHOD FOR FUSING RADAR/CAMERA OBJECT DATA AND LiDAR SCAN POINTS
CN105745122A (zh) * 2013-09-23 2016-07-06 大众汽车有限公司 用于显示车辆周围环境的驾驶员辅助系统
CN106585623A (zh) * 2016-12-21 2017-04-26 驭势科技（北京）有限公司 一种用于探测车辆周围目标的探测系统以及其应用
CN106707293A (zh) * 2016-12-01 2017-05-24 百度在线网络技术（北京）有限公司 用于车辆的障碍物识别方法和装置
CN206348459U (zh) * 2016-11-23 2017-07-21 岭纬科技有限公司 基于多传感器融合的三维视觉传感装置
CN108303711A (zh) * 2017-12-19 2018-07-20 深圳市海梁科技有限公司 一种反光条及智能汽车激光雷达检测系统

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106646494A (zh) * 2016-11-03 2017-05-10 上海博未传感技术有限公司 一种采用发射和接收光路复用结构的激光雷达系统

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160018524A1 (en) * 2012-03-15 2016-01-21 GM Global Technology Operations LLC SYSTEM AND METHOD FOR FUSING RADAR/CAMERA OBJECT DATA AND LiDAR SCAN POINTS
CN102779280A (zh) * 2012-06-19 2012-11-14 武汉大学 一种基于激光传感器的交通信息提取方法
CN102837658A (zh) * 2012-08-27 2012-12-26 北京工业大学 一种智能车辆中多激光雷达数据融合系统和方法
CN105745122A (zh) * 2013-09-23 2016-07-06 大众汽车有限公司 用于显示车辆周围环境的驾驶员辅助系统
CN104899855A (zh) * 2014-03-06 2015-09-09 株式会社日立制作所 三维障碍物检测方法和装置
CN206348459U (zh) * 2016-11-23 2017-07-21 岭纬科技有限公司 基于多传感器融合的三维视觉传感装置
CN106707293A (zh) * 2016-12-01 2017-05-24 百度在线网络技术（北京）有限公司 用于车辆的障碍物识别方法和装置
CN106585623A (zh) * 2016-12-21 2017-04-26 驭势科技（北京）有限公司 一种用于探测车辆周围目标的探测系统以及其应用
CN108303711A (zh) * 2017-12-19 2018-07-20 深圳市海梁科技有限公司 一种反光条及智能汽车激光雷达检测系统

Cited By (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112327308A (zh) * 2019-07-19 2021-02-05 阿里巴巴集团控股有限公司 物体检测方法、装置、系统及设备
CN111144211A (zh) * 2019-08-28 2020-05-12 华为技术有限公司 点云显示方法和装置
CN111144211B (zh) * 2019-08-28 2023-09-12 华为技术有限公司 点云显示方法和装置
EP3998548A4 (en) * 2019-08-28 2022-11-09 Huawei Technologies Co., Ltd. METHOD AND APPARATUS FOR DOT RECRUDESCENCE DISPLAY
CN112445215A (zh) * 2019-08-29 2021-03-05 阿里巴巴集团控股有限公司 自动导引车行驶控制方法、装置及计算机系统
CN112558103A (zh) * 2019-09-10 2021-03-26 奥迪股份公司 障碍物检测设备、障碍物的检测方法及装置、计算机设备
CN111443360A (zh) * 2020-04-20 2020-07-24 北京易控智驾科技有限公司 矿区无人驾驶系统道路边界自动采集装置和识别方法
CN111443360B (zh) * 2020-04-20 2022-06-24 北京易控智驾科技有限公司 矿区无人驾驶系统道路边界自动采集装置和识别方法
CN112348777A (zh) * 2020-10-19 2021-02-09 深圳市优必选科技股份有限公司 人体目标的检测方法、装置及终端设备
CN112348777B (zh) * 2020-10-19 2024-01-12 深圳市优必选科技股份有限公司 人体目标的检测方法、装置及终端设备
CN112415538A (zh) * 2020-11-16 2021-02-26 重庆长安汽车股份有限公司 自动驾驶车辆绕行锥形筒的规划方法、系统及车辆
CN112485774B (zh) * 2020-11-26 2024-03-15 中国第一汽车股份有限公司 一种车载激光雷达标定方法、装置、设备及存储介质
CN112485774A (zh) * 2020-11-26 2021-03-12 中国第一汽车股份有限公司 一种车载激光雷达标定方法、装置、设备及存储介质
CN113095266A (zh) * 2021-04-19 2021-07-09 北京经纬恒润科技股份有限公司 一种角度识别方法、装置及设备
CN113095266B (zh) * 2021-04-19 2024-05-10 北京经纬恒润科技股份有限公司 一种角度识别方法、装置及设备
EP4091424A1 (en) * 2021-05-19 2022-11-23 AGCO International GmbH Residue spread monitoring
CN113467450A (zh) * 2021-07-01 2021-10-01 北京小狗吸尘器集团股份有限公司 无人机器控制方法、装置、计算机设备及存储介质
CN113985389A (zh) * 2021-09-30 2022-01-28 苏州浪潮智能科技有限公司 一种时间同步校准装置、自动识别路径设备、方法及介质
CN113985389B (zh) * 2021-09-30 2024-02-09 苏州浪潮智能科技有限公司 一种时间同步校准装置、自动识别路径设备、方法及介质
CN115327571A (zh) * 2022-07-29 2022-11-11 武汉理工大学 一种基于平面激光雷达的三维环境障碍物检测系统及方法
CN116412814A (zh) * 2023-06-12 2023-07-11 旷智中科(北京)技术有限公司 一种基于激光雷达的图像构建导航辅助系统

Also Published As

Publication number Publication date
CN108318895A (zh) 2018-07-24
CN108318895B (zh) 2020-02-07

Similar Documents

Publication Publication Date Title
CN108318895B (zh) 用于无人驾驶车辆的障碍物识别方法、装置及终端设备
US11027653B2 (en) Apparatus, system and method for preventing collision
US20210122364A1 (en) Vehicle collision avoidance apparatus and method
US20230068113A1 (en) Sensor calibration facility
KR102614323B1 (ko) 수동 및 능동 측정을 이용한 장면의 3차원 지도 생성
JP6860656B2 (ja) 車両の形状に適応したダイナミックステアドｌｉｄａｒ
KR102201290B1 (ko) 차량용 디스플레이 장치 및 차량
US9863775B2 (en) Vehicle localization system
US20240142607A1 (en) Information processing device, information processing method, computer program, and mobile device
US11858493B2 (en) Method of sharing and using sensor data
US20220179094A1 (en) Systems and methods for implementing a tracking camera system onboard an autonomous vehicle
CN111736153A (zh) 用于无人驾驶车辆的环境检测系统、方法、设备和介质
EP4102251A1 (en) Determination of atmospheric visibility in autonomous vehicle applications
EP4170606A1 (en) Identification of real and image sign detections in driving applications
US20190236989A1 (en) Interactive Projection System
CN113093176B (zh) 线状障碍物检测方法、装置、电子设备和存储介质
CN211032395U (zh) 自动驾驶车辆
US20240036212A1 (en) Lane boundary detection using sub-short range active light sensor
WO2023221118A1 (zh) 信息处理方法及装置、电子设备及存储介质
US20230267746A1 (en) Information processing device, information processing method, and program
US12099124B1 (en) Integrated sensor assembly
Meydani State-of-the-Art Analysis of the Performance of the Sensors Utilized in Autonomous Vehicles in Extreme Conditions
US11567173B2 (en) Systems and methods for increasing lidar sensor coverage
US20240174173A1 (en) Notification device
WO2022160101A1 (zh) 朝向估计方法、装置、可移动平台及可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17935177

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 22/09/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17935177

Country of ref document: EP

Kind code of ref document: A1