CN114064365A - Multi-sensor data stream backup system and method - Google Patents
Multi-sensor data stream backup system and method Download PDFInfo
- Publication number
- CN114064365A CN114064365A CN202111421686.8A CN202111421686A CN114064365A CN 114064365 A CN114064365 A CN 114064365A CN 202111421686 A CN202111421686 A CN 202111421686A CN 114064365 A CN114064365 A CN 114064365A
- Authority
- CN
- China
- Prior art keywords
- unit
- data
- sensor
- data stream
- computing
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/14—Error detection or correction of the data by redundancy in operation
- G06F11/1402—Saving, restoring, recovering or retrying
- G06F11/1446—Point-in-time backing up or restoration of persistent data
- G06F11/1448—Management of the data involved in backup or backup restore
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/54—Interprogram communication
- G06F9/546—Message passing systems or structures, e.g. queues
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a multi-sensor data stream backup system and a method, wherein the system comprises: the device comprises a sensor unit, a calculation unit, a data distribution unit, a monitoring unit and a switching unit; the sensor unit is connected with the data distribution unit and used for transmitting the acquired sensor data stream to the data distribution unit, the switching unit is connected between the data distribution unit and the calculation unit, the monitoring unit is respectively connected with the calculation unit and the switching unit and used for monitoring the state of the calculation unit and controlling the switching unit to switch the configuration channel between the data distribution unit and the calculation unit according to the state of the calculation unit, the data distribution unit is used for distributing the corresponding sensor data stream to the calculation unit through the communicated configuration channel based on a data backup strategy, and the calculation unit processes the received sensor data. The invention distributes and processes the sensor data first, which can effectively reduce the number of sensors or the data transmission cost of the computing resource.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to a multi-sensor data stream backup system and a method.
Background
At present, most high-level automatic driving computing platforms carry 2 or more computing units, and are provided with a plurality of cameras, laser radars, millimeter wave radars and other sensors.
The design of a common vehicle-mounted computing platform is that a sensor is matched with a computing unit point to point. Due to the change of the algorithm scheme, data interaction of different sensors needs to be transmitted through a channel between the computing units.
The upper-level application and algorithm of automatic driving are iterated rapidly, and the scheme of the external access sensor can change at any time. The existing scheme has poor sensor reusability, needs to configure a plurality of sets of identical sensors, and undoubtedly increases the cost and the complexity of the layout of the sensors. In addition, another solution is to transmit the data of different sensors through channels between the computing units. But adds additional overhead to the computing resources and is not cost effective for the otherwise scarce computing resources. And can introduce complexity to the development of the computing unit software.
Disclosure of Invention
In view of this, the present invention provides a multi-sensor data stream backup system and method, which can effectively reduce the number of sensors or the data transmission cost of computing resources by allocating and then processing sensor data.
A first aspect of the present invention provides a multi-sensor data stream backup system, comprising: the device comprises a sensor unit, a calculation unit, a data distribution unit, a monitoring unit and a switching unit; the sensor unit is connected with the data distribution unit and used for transmitting acquired sensor data streams to the data distribution unit, the switching unit is connected between the data distribution unit and the calculation unit, the monitoring unit is respectively connected with the calculation unit and the switching unit and used for monitoring the state of the calculation unit and controlling the switching unit to switch a configuration channel between the data distribution unit and the calculation unit according to the state of the calculation unit, the data distribution unit is used for distributing the corresponding sensor data streams to the calculation unit through the communicated configuration channel based on a data backup strategy, and the calculation unit processes the received sensor data.
Further, the number of the computing units is one or more.
Further, the data distribution unit comprises a data exchange module and a data deserializing module;
the input end of the data exchange module is connected with the sensor unit, and the output end of the data exchange module is connected with the computing unit and is used for receiving the corresponding sensor data stream transmitted by the sensor unit and directly transmitting the corresponding sensor data stream to the computing unit based on a data backup strategy;
the input end of the data string reducing module is connected with the sensor unit, the output end of the data string reducing module is connected with the computing unit, the switching unit is connected between the data string reducing module and the computing unit, and when the monitoring unit controls the switching unit to switch so that the configuration channels between the data string reducing module and the computing unit are communicated, the data string reducing module transmits corresponding sensor data streams to the computing unit through the communicated configuration channels based on a data backup strategy.
Further, the switching unit includes a switch, and the switch is connected between the data deserializing module and the computing unit and is configured to switch a configuration channel between the data deserializing module and the computing unit.
Further, the sensor unit is at least one of a camera, a laser radar sensor and a millimeter wave radar sensor.
Further, the sensor unit comprises a camera, the camera is connected with the data deserializing module, and when the monitoring unit controls the switching unit to switch to enable the configuration channel between the data deserializing module and the computing unit to be communicated, the data deserializing module transmits the data stream collected by the camera to the computing unit based on a data backup strategy.
Further, the sensor unit includes a radar sensor, the radar sensor is connected to the data exchange module, and the data exchange module directly transmits the data stream collected by the radar sensor to the computing unit based on a data backup strategy.
A second aspect of the present invention provides a method for backing up multiple sensor data streams, where the method is implemented based on the sensor data stream backup system described above, and the method includes: the sensor unit collects sensor data streams and transmits the collected sensor data streams to the data distribution unit; the monitoring unit monitors the state of the computing unit and controls the switching unit to switch the configuration channel between the data distribution unit and the computing unit according to the state of the computing unit; and the data distribution unit distributes corresponding sensor data streams to the calculation unit based on a data backup strategy, and the calculation unit processes the received sensor data.
Further, when the monitoring unit controls the switching unit to switch to enable the configuration channel between the data deserializing module in the data distribution unit and the computing unit to be communicated, the data distribution unit transmits the data stream acquired by the camera in the sensor unit to the computing unit through the communicated configuration channel based on a data backup strategy.
Further, the data exchange module in the data distribution unit directly transmits the data stream collected by the radar sensor in the sensor unit to the computing unit based on a data backup strategy.
According to the sensor data stream backup system and the method, the sensor data is distributed and then processed, so that the number of sensors or the data transmission cost of computing resources can be effectively reduced.
Drawings
For purposes of illustration and not limitation, the present invention will now be described in accordance with its preferred embodiments, particularly with reference to the accompanying drawings, in which:
FIG. 1 is a schematic diagram of a prior art in-vehicle computing platform;
FIG. 2 is a schematic structural diagram of a multi-sensor data stream backup system according to an embodiment of the present invention;
FIG. 3 is an architecture diagram of a multi-sensor data stream backup system according to an embodiment of the present invention;
fig. 4 is a flowchart of a multi-sensor data stream backup method according to a second embodiment of the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a detailed description of the present invention will be given below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and features of the embodiments may be combined with each other without conflict.
In the following description, numerous specific details are set forth to provide a thorough understanding of the present invention, and the described embodiments are merely a subset of the embodiments of the present invention, rather than a complete embodiment. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
Fig. 1 is a schematic structural diagram of a conventional in-vehicle computing platform. As shown in fig. 1, the existing in-vehicle computing platform is designed such that the sensors are matched point-to-point with the computing unit. Due to the change of the algorithm scheme, data interaction of different sensors needs to be transmitted through a channel between the computing units.
Fig. 2 is a schematic structural diagram of a multi-sensor data stream backup system according to an embodiment of the present invention. The multi-sensor data stream backup system is used for carrying out allocation and post-processing on the sensor data, so that the number of sensors can be effectively reduced, or the data transmission cost of computing resources can be reduced.
Referring to fig. 2, the multi-sensor data stream backup system 100 includes a sensor unit 101, a data distribution unit 102, a first calculation unit 103, a second calculation unit 104, a monitoring unit 105, and a switching unit 106. The sensor data stream collected by the sensor unit 101 firstly enters the data distribution unit 102, the data distribution unit 102 distributes the sensor data to the first calculation unit 103 and/or the second calculation unit 104, the monitoring unit 105 controls the switching unit 106 to switch the configuration channel of the data distribution unit 102 to the sensor unit 101 according to the states of the first calculation unit 103 and the second calculation unit 104, and the corresponding sensor data stream is processed by the corresponding first calculation unit 103 and/or the second calculation unit 104.
The sensor unit 101 is configured to collect a sensor data stream. The sensor data stream includes, among other things, an image data stream, a point cloud data stream, and so on.
The data distribution unit 102 is connected to the sensor unit 101, and configured to receive the sensor data stream collected by the sensor unit 101, and distribute the corresponding sensor data stream to the first computing unit 103 and/or the second computing unit 104 based on the data backup policy.
The first computing unit 103 is connected to the data distribution unit 102, and is configured to receive the corresponding sensor data stream distributed by the data distribution unit 102 and process the received corresponding sensor data stream.
The second computing unit 104 is connected to the data distribution unit 102, and is configured to receive the corresponding sensor data stream distributed by the data distribution unit 102 and process the received corresponding sensor data stream.
The monitoring unit 105 is connected to the first calculating unit 103 and the second calculating unit 104, and is configured to monitor states of the first calculating unit 103 and the second calculating unit 104.
The switching unit 106 is connected between the data distribution unit 102 and the first and second calculation units 103 and 104, and the switching unit 106 is further connected with the monitoring unit 105, and is used for switching the configuration channel communicated between the data distribution unit 102 and the first and second calculation units 103 and 104 according to the state data of the first and second calculation units 103 and 104 monitored by the monitoring unit 105.
The data distribution unit 102, the monitoring unit 105 and the switching unit 106 are added in the sensor data stream backup system, so that the first computing unit 103 and the second computing unit 104 process different sensor data streams, the monitoring unit 105 monitors the states of the first computing unit 103 and the second computing unit 104 in real time, the switching unit 106 is controlled to switch the configuration channel of the sensor unit 101 according to the states and requirements of the first computing unit 103 and the second computing unit 104, and the first computing unit 103 and/or the second computing unit 104 process different sensor data streams.
When the above-mentioned sensor data stream backup system is used, the sensor data stream collected by the sensor unit 101 first enters the data distribution unit 102, and the data distribution unit 102 distributes the corresponding sensor data to the first computing unit 103 and/or the second computing unit 104 based on the data stream backup policy. The monitoring unit 105 controls the switching unit 106 to switch the configuration channels of the data distribution unit 102 to the sensor unit 101 according to the state or algorithm requirements of the first computing unit 103 and the second computing unit 104, and the first computing unit 103 and/or the second computing unit 104 processes the corresponding sensor data streams.
In the present embodiment, the sensor unit 101 may be a camera, a laser radar, a millimeter wave radar, or the like. Image data streams, point cloud data streams and the like are acquired through sensors such as cameras, laser radars, millimeter wave radars and the like.
Since the camera has an independent configuration channel, specifically, the first computing unit 103 or the second computing unit 104 processes the image data stream collected by the camera, the monitoring unit 105 needs to control the switching unit 106 to switch the configuration channel between the camera, the data distribution unit 102 and the first computing unit 103 or the second computing unit 104 according to the state or algorithm requirement of the first computing unit 103 or the second computing unit 104. When the monitoring unit 105 controls the switching unit 106 to switch the communication of the configuration channels between the camera and the data distribution unit 102 and the first calculation unit 103, the sensor data collected by the camera can be transmitted to the first calculation unit 103 through the data distribution unit 102, and the image data stream collected by the camera is processed by the first calculation unit 103. When the monitoring unit 105 controls the switching unit 106 to switch the communication of the camera with the configuration channels between the data distribution unit 102 and the first computing unit 104, the sensor data collected by the camera can be transmitted to the second computing unit 104 through the data distribution unit 102, and the image data stream collected by the camera is processed by the second computing unit 104.
In some embodiments, the switching unit 106 includes a switch, and the monitoring unit 105 controls the switch to switch, so that the first calculating unit 103 or the second calculating unit 104 is in communication with the data distribution unit 102.
As shown in fig. 3, in this embodiment, the data distribution unit 102 includes a data exchange module 1021 and a data deserializing module 1022, input ends of the data exchange module 1021 and the data deserializing module 1022 are connected to the sensor unit 101, output ends of the data exchange module 1021 and the data deserializing module 1022 are further connected to the first computing unit 103 and the second computing unit 104, respectively, sensor data collected by a sensor of an ethernet interface such as a lidar is distributed to the first computing unit 103 and/or the second computing unit 104 through the data exchange module 1021, and sensor data collected by a camera is distributed to the first computing unit 103 and/or the second computing unit 104 through the data deserializing module 1022. One end of the switch is connected to the monitoring unit 105, the other end of the switch is connected to the data string reducing module 1022, the monitoring unit 105 controls the switch to switch, the first computing unit 103 or the second computing unit 104 is communicated with the data string reducing module 1022, and the first computing unit 103 or the second computing unit 104 processes sensor data collected by the camera.
In this embodiment, the switch uses a triode, and the switching of the triode is controlled by the monitoring unit 105, so that the first computing unit 103 and the second computing unit 104 are communicated with the data deserializing module 1022.
In some embodiments, data exchange module 1021 may employ an 88EA6321 exchange chip. The data deserializing module 1021 may employ a MAX96712 chip.
In this embodiment, the monitoring unit 105 is a vehicle-regulated MCU, and monitors the states of the first calculating unit 103 and the second calculating unit 104 through the vehicle-regulated MCU, and the vehicle-regulated MCU controls the switches to communicate the first calculating unit 103 and the second calculating unit 104 with the data distributing unit 102. The vehicle-scale MCU can adopt a TC397 chip, the TC397 chip monitors the state of the first calculating unit 103 or the second calculating unit 104, and switching is performed through a control triode.
According to the invention, the data distribution unit 102 distributes the corresponding sensor data streams to the calculation unit according to the data stream backup strategy, so that the configuration number of the sensors is effectively reduced; the application also adds a monitoring unit and a change-over switch, and can realize the configuration and management of any computing unit.
Fig. 4 is a flowchart of a multi-sensor data stream backup method according to a second embodiment of the present invention. The sensor data stream backup method is used for carrying out allocation and post-processing on the sensor data, so that the number of sensors can be effectively reduced, or the data transmission cost of computing resources can be reduced. The method is implemented based on the sensor data stream backup system as described above.
Referring to fig. 4, the method for backing up a multi-sensor data stream includes the following steps:
s1, the sensor unit collects the sensor data stream and transmits the collected sensor data stream to the data distribution unit.
And S2, the monitoring unit monitors the states of the first computing unit and the second computing unit, and the switching unit is controlled to switch the configuration channel communicated between the data distribution unit and the first computing unit and/or the second computing unit.
S3, the data allocation unit allocates the corresponding sensor data stream to the first computing unit and/or the second computing unit based on the data backup policy.
And S4, the first computing unit and/or the second computing unit receives the corresponding sensor data stream distributed by the data distribution unit and processes the received corresponding sensor data stream.
In this embodiment, sensor data collected by a sensor of an ethernet interface such as a lidar is directly distributed to the first computing unit and/or the second computing unit through the data exchange module in the data distribution unit, and the received corresponding sensor data stream is processed by the first computing unit and/or the second computing unit.
When the monitoring unit controls the switching unit to switch the communication of the configuration channels between the data string reducing module in the data distribution unit and the first calculation unit and/or the second calculation unit, the data string reducing module in the data distribution unit distributes the sensor data collected by the camera to the first calculation unit and/or the second calculation unit, and the first calculation unit and/or the second calculation unit process the received corresponding sensor data streams.
According to the multi-sensor data stream backup method, the corresponding sensor data streams are distributed to the computing unit through the data distribution unit according to the data stream backup strategy, so that the configuration number of the sensors is effectively reduced; and a monitoring unit and a change-over switch are additionally arranged, so that the configuration and management of the sensor of any computing unit can be realized.
The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A multi-sensor data stream backup system, comprising: the device comprises a sensor unit, a calculation unit, a data distribution unit, a monitoring unit and a switching unit;
the sensor unit is connected with the data distribution unit and used for transmitting acquired sensor data streams to the data distribution unit, the switching unit is connected between the data distribution unit and the calculation unit, the monitoring unit is respectively connected with the calculation unit and the switching unit and used for monitoring the state of the calculation unit and controlling the switching unit to switch a configuration channel between the data distribution unit and the calculation unit according to the state of the calculation unit, the data distribution unit is used for distributing the corresponding sensor data streams to the calculation unit through the communicated configuration channel based on a data backup strategy, and the calculation unit processes the received sensor data.
2. The multi-sensor data stream backup system according to claim 1, wherein the number of computing units is one or more.
3. The multi-sensor data stream backup system of claim 1, wherein the data distribution unit comprises a data exchange module and a data deserialization module;
the input end of the data exchange module is connected with the sensor unit, and the output end of the data exchange module is connected with the computing unit and is used for receiving the corresponding sensor data stream transmitted by the sensor unit and directly transmitting the corresponding sensor data stream to the computing unit based on a data backup strategy;
the input end of the data string reducing module is connected with the sensor unit, the output end of the data string reducing module is connected with the computing unit, the switching unit is connected between the data string reducing module and the computing unit, and when the monitoring unit controls the switching unit to switch so that the configuration channels between the data string reducing module and the computing unit are communicated, the data string reducing module transmits corresponding sensor data streams to the computing unit through the communicated configuration channels based on a data backup strategy.
4. The multi-sensor data stream backup system according to claim 3, wherein the switching unit comprises a switch connected between the data deserializing module and the computing unit for switching a configuration channel between the data deserializing module and the computing unit.
5. The multi-sensor data stream backup system of claim 1, wherein said sensor unit is at least one of a camera, a lidar sensor, and a millimeter-wave radar sensor.
6. The multi-sensor data stream backup system according to claim 3, wherein the sensor unit comprises a camera, the camera is connected to the data deserializing module, and when the monitoring unit controls the switching unit to switch the configuration channel between the data deserializing module and the computing unit to be communicated, the data deserializing module transmits the data stream collected by the camera to the computing unit based on a data backup strategy.
7. The multi-sensor data stream backup system according to claim 3, wherein the sensor unit comprises a radar sensor, the radar sensor is connected to the data exchange module, and the data exchange module directly transmits the data stream collected by the radar sensor to the computing unit based on a data backup strategy.
8. A multi-sensor data stream backup method implemented based on the multi-sensor data stream backup system of any one of claims 1 to 7, the method comprising:
the sensor unit collects sensor data streams and transmits the collected sensor data streams to the data distribution unit;
the monitoring unit monitors the state of the computing unit and controls the switching unit to switch the configuration channel between the data distribution unit and the computing unit according to the state of the computing unit;
and the data distribution unit distributes corresponding sensor data streams to the calculation unit based on a data backup strategy, and the calculation unit processes the received sensor data.
9. The multi-sensor data stream backup method according to claim 8, wherein when the monitoring unit controls the switching unit to switch to connect the configuration channel between the data deserializing module in the data distribution unit and the computing unit, the data distribution unit transmits the data stream collected by the camera in the sensor unit to the computing unit through the connected configuration channel based on a data backup strategy.
10. The multi-sensor data stream backup method according to claim 8, wherein a data exchange module in the data distribution unit directly transmits the data stream collected by the radar sensor in the sensor unit to the computing unit based on a data backup strategy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111421686.8A CN114064365A (en) | 2021-11-26 | 2021-11-26 | Multi-sensor data stream backup system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111421686.8A CN114064365A (en) | 2021-11-26 | 2021-11-26 | Multi-sensor data stream backup system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114064365A true CN114064365A (en) | 2022-02-18 |
Family
ID=80276690
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111421686.8A Pending CN114064365A (en) | 2021-11-26 | 2021-11-26 | Multi-sensor data stream backup system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114064365A (en) |
-
2021
- 2021-11-26 CN CN202111421686.8A patent/CN114064365A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105518620A (en) | Network card configuration method and resource management center | |
| CN111158900B (en) | Lightweight distributed parallel computing system and method | |
| CN101534510B (en) | Method and system for sharing distributed frequency spectrum | |
| CN211427332U (en) | VPX integrated circuit board based on FPGA array | |
| CN114802058A (en) | Intelligent electric vehicle regional architecture vehicle-mounted networked control system and scheduling method | |
| CN111835849A (en) | Method and device for enhancing service capability of access network | |
| CN116166434A (en) | Processor allocation method and system, device, storage medium and electronic equipment | |
| CN112020001A (en) | Time slot resource allocation method for multi-station multi-machine system of unmanned aerial vehicle | |
| CN111665492A (en) | Airborne distributed comprehensive radio frequency sensor system | |
| CN110932920A (en) | Network topology structure | |
| CN116582173B (en) | Method, device and storage medium for processing data by satellite-based distributed network | |
| CN112511621B (en) | Data transmission method and device, storage medium, and electronic device | |
| CN114064365A (en) | Multi-sensor data stream backup system and method | |
| CN110069430B (en) | Data acquisition system of chrysanthemum chain structure and self-adaptive transmission method | |
| CN111262723A (en) | Edge intelligent computing platform based on modularized hardware and software definition | |
| CN113608861B (en) | Virtualized distribution method and device for software load computing resources | |
| JP2012257249A (en) | System integrating command and control system and network management system | |
| CN112148663A (en) | Data exchange chip and server | |
| CN114257467A (en) | Low-power-consumption CAN bus communication network based on dynamic management and control method thereof | |
| CN109067631B (en) | Multi-node automobile electric control method based on real-time Ethernet communication protocol architecture | |
| CN111615082A (en) | Heterogeneous automatic driving vehicle formation calculation task collaborative execution method | |
| CN113993212A (en) | Networking method of scattering communication, central station, peripheral station and scattering communication system | |
| US10477544B2 (en) | System and method for sharing RF signal processing | |
| CN116149861B (en) | High-speed communication method based on VPX structure | |
| CN102104579B (en) | System and method for reducing trans-frame media stream resources |
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 |