Adaptive function implementation method of laser radar
The adaptive LiDAR achieves adaptability under different network locations and time synchronization protocols, solving the adaptability problem of LiDAR on the mounting platform, ensuring the accuracy and continuity of data timestamps, and improving the stability and data processing efficiency of the system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG FREE OPTICAL TECH CO LTD
- Filing Date
- 2022-09-22
- Publication Date
- 2026-06-09
AI Technical Summary
Existing lidar systems have poor adaptability to different platforms, making it difficult to meet the requirements for time synchronization and data processing under different conditions and network locations.
A method for implementing adaptive functionality in lidar is provided, including network location adaptation, time synchronization protocol type adaptation, time synchronization function adaptation, and timestamp-corresponding measurement time adaptation. Time synchronization is achieved through NTP and PTP time synchronization network protocols, and absolute and relative timestamps are added to the data to ensure data continuity.
It improves the adaptability of lidar under different master control systems and network locations, ensures the accuracy and continuity of data timestamps, and reduces the impact of time synchronization delays and out-of-order issues.
Smart Images

Figure CN115656980B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of lidar technology, and more specifically to a method for implementing the adaptive function of lidar. Background Technology
[0002] LiDAR, as a ranging sensor, is widely used in robotics and AGVs. The main control system of a robot or AGV equipped with LiDAR needs to coordinate the actions of multiple sensors, including the LiDAR, requiring time synchronization of the LiDAR and other sensors. To facilitate the fusion of multi-sensor data, timestamps reflecting the measurement time need to be added to the measurement data from the LiDAR and other sensors. Time synchronization between sensors effectively reduces time differences caused by physical hardware and software settings between LiDAR sensors and between LiDAR and other types of sensors. The introduction of data timestamps effectively eliminates the negative impacts caused by delays and out-of-order transmission during data transmission.
[0003] The main control system of the robot or AGV, along with LiDAR and other types of sensors, constitute a local area network. Within this local area network, the LiDAR is mostly located at the terminal position, and time synchronization can be achieved through the network.
[0004] There are three existing collaboration modes between lidar and its platform's main control system: 1. No time synchronization between the lidar and the main control system. The lidar data is timestamped, and the main control system processes the lidar data based on the timestamped data. There are no hardware or software requirements for time synchronization with the lidar itself. The main control system can only passively receive and process data. The stability of the lidar clock and the accuracy of the base time setting directly determine the accuracy of the system data. 2. Synchronization between the lidar and the main control system via I / O signals. An additional I / O data line connection is required between the lidar and the main control system. The lidar sends a high-level I / O signal at a specific frequency or at a specific time (such as the start or end time of each scan). The main control system uses the received high-level I / O signal as the time reference for data packets or frame data. This synchronization method requires low latency in transmitting and receiving I / O signals and high real-time performance. However, due to limitations in the actual hardware conditions of the radar and central control system, factors such as loose connections, external interference, and time delays caused by non-parallel tasks result in time synchronization falling far short of design expectations. 3. The radar and the main control system have time synchronization. The main control system is configured with stable synchronization service resources. The radar, as a terminal, periodically synchronizes with the main control system. The main control system is specifically designed for specific tasks and deterministic control loops. Sensors such as the radar in the system must meet the condition of time accuracy. Any synchronization failure (hardware or software) will cause abnormal system data processing, leading to inefficient or invalid use of radar data. The spatial reconstruction, navigation, and positioning functions of the main control system will be directly affected. This collaborative mode places strict requirements on the hardware and software configuration and operational stability of the radar, the main control system, and the connection between them.
[0005] In real-world applications, the status of a central control system equipped with lidar is complex and variable due to external influences, and the networking requirements are also diverse.
[0006] The radar platform can exist in multiple states, including having time synchronization service, not having synchronization service, or having synchronization service intermittently. When synchronization service is available, there are differences between single-server and multi-server states, requiring the lidar to meet the data application needs under various states.
[0007] Radar platforms typically connect multiple sensor modules, forming a small local area network (LAN) composed of a central control system and the various sensor modules. Depending on the actual needs, the lidar may be required to be at the end of the network or as an intermediate node. Therefore, the lidar must meet the hardware and software configuration requirements for different network positions. Summary of the Invention
[0008] The purpose of this invention is to provide a method for implementing the adaptive function of a lidar, so as to solve the problem of poor adaptability of existing lidars to the lidar mounting platform.
[0009] To solve the above technical problems, the present invention provides a technical solution: a method for implementing the adaptive function of a lidar: the adaptive function of the lidar includes network location adaptation, time synchronization protocol type adaptation, time synchronization function adaptation, and timestamp corresponding to measurement time adaptation;
[0010] The adaptive implementation method for network location and time synchronization protocol type is as follows: the LiDAR used is applicable to at least two time synchronization networks, and simultaneously includes a network time protocol server and a network time protocol client corresponding to each time synchronization network; when the LiDAR acts as a client, it sends time synchronization request messages for all types of time synchronization networks through its communication port, determines the corresponding time synchronization network type based on the received reply messages, and calibrates the local clock; when the LiDAR acts as a server, it determines the corresponding time synchronization network type based on the received time synchronization request messages, and sends reply messages using the network time protocol server corresponding to that time synchronization network type.
[0011] The adaptive implementation method of time synchronization function is as follows: the LiDAR used is connected to the time synchronization server through a local area network. The time synchronization server calibrates the LiDAR according to a certain standard protocol. The LiDAR adds a relative timestamp generated based on the uncalibrated local clock and an absolute timestamp generated based on the calibrated local clock to the LiDAR data. The absolute timestamp contains a specific flag bit used to indicate whether the time synchronization server is in a continuous synchronization state with the LiDAR.
[0012] The adaptive implementation method of timestamp corresponding to measurement time is as follows: obtain the initial reference time T0 when the light emission command is sent to the radar transmission control module and the time register module, the light emission time T1 of the lidar, and the time when the laser illuminates the target being measured T2. Obtain the delay duration based on the time interval between the above adjacent times, and then assign a timestamp to the packaged data generated by the lidar based on T0 and the delay duration.
[0013] According to the above scheme, the time synchronization networks applicable to lidar include NTP time synchronization networks and PTP time synchronization networks.
[0014] According to the above scheme, the time synchronization server calibrates the lidar time according to the IEEE1588 standard.
[0015] According to the above scheme, the time period for the time synchronization server to synchronize the LiDAR is set according to user needs.
[0016] According to the above scheme, after the time synchronization server successfully synchronizes the LiDAR, if the LiDAR does not receive a reply message from the time synchronizer in a certain instance, the LiDAR will use the clock information that was calibrated when it last received a reply message from the time synchronizer.
[0017] An adaptive lidar includes a PHY chip and an FPGA chip;
[0018] Based on the PHY chip, network time protocol servers and network time protocol clients were established for each time synchronization network of the lidar, as well as a TCP / IP module for network communication with external systems.
[0019] Based on the FPGA chip, a crystal oscillator, timing module, and time synchronization module were established for time synchronization; a time calculation module and distance calculation module for data processing; a time register module and data packing module for data and integration; and a light output and time register synchronization module and a transmission control module for light output control.
[0020] The beneficial effects of this invention are: the method enables the lidar to have adaptive functions in terms of network location, time synchronization protocol type, time synchronization function, and timestamp corresponding to measurement time, thereby improving the adaptability of the lidar when it is mounted on different main control systems. Attached Figure Description
[0021] Figure 1 This is a schematic diagram of a time synchronization adaptive process according to an embodiment of the present invention;
[0022] Figure 2 This is a schematic diagram of an adaptive process for timestamps corresponding to measurement times according to an embodiment of the present invention. Detailed Implementation
[0023] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this disclosure. All other embodiments obtained by those skilled in the art based on the described embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.
[0024] This invention provides a lidar with adaptive functionality. The adaptive functionality is manifested in four aspects: first, network location adaptation; second, time synchronization protocol type adaptation; third, time synchronization function adaptation; and fourth, timestamp-to-measurement-time adaptation.
[0025] Adaptive network location and time synchronization protocols:
[0026] LiDAR is suitable for both NTP and PTP time synchronization networks. It includes both a Network Time Protocol (NTP) server and a NTP client. On its communication port, the LiDAR acts as a server, receiving synchronization request messages and sending response messages; as a client, it sends synchronization request messages outwards and also receives response messages from other servers.
[0027] The specific implementation method is as follows:
[0028] 1. Implementation based on main control chip and PHY chip
[0029] 1) Based on the Ethernet TCP network protocol, the current clock inside the radar can be configured / modified through this protocol.
[0030] 2) Implement a Network Time Protocol (NTP) client based on Ethernet UDP communication that conforms to both NTP and PTP.
[0031] 3) Implement a Network Time Protocol (NTP) server based on Ethernet UDP communication that conforms to both NTP and PTP.
[0032] 2. Determine which network time protocol (NTP or PTP) the radar uses through built-in parameter settings or external user configuration, and confirm whether the corresponding client and server are enabled. The radar is configured by default to have both time synchronization services on the network simultaneously, and all types of clients and servers must be enabled at the same time.
[0033] 3. In the default settings, the radar simultaneously sends NTP time synchronization request messages and PTP time synchronization request messages. Based on the received message format, it determines which synchronization protocol is currently available on the network and calibrates the local clock based on the reply message.
[0034] 4. In the default settings, the radar simultaneously enables two types of synchronization servers. It determines whether the received synchronization service request message is NTP or PTP based on its format, and then provides a reply message based on the corresponding server.
[0035] Time synchronization function adaptive:
[0036] LiDAR relies on local area network (LAN) time synchronization services. It synchronizes the radar time using a network time protocol, adding absolute and relative timestamps to the radar measurement data. The timestamps vary depending on changes in the external system's time synchronization service status. Regardless of the situation, the LiDAR provides measurement data, and its packaged data includes the most optimal timestamp information based on the current external environment, ensuring continuous availability of radar data. Simultaneously, external systems can determine their own operational status based on the number of timestamps and timestamp identifier information received from the LiDAR data.
[0037] The specific implementation method is as follows:
[0038] The lidar is physically connected to the system. The lidar sends a time synchronization request based on the system's IP address or IP address range (or a user-configured IP address) and seeks a response from the time synchronization server.
[0039] 1. If the radar sends k (based on user settings) time synchronization requests from the moment it transitions to normal operation after initialization, and fails to receive a synchronization message from the time synchronization server, it is assumed that a time synchronization server does not exist within the current main control system. The radar then provides a relative timestamp for its data based on its own clock, along with an absolute timestamp containing a specific flag and an empty time information. This flag indicates that the radar has not received time synchronization service since its startup.
[0040] 2. Upon receiving a synchronization message from the time synchronization server, time calibration is performed according to the IEEE 1588 standard. The calibrated local clock is then used to provide absolute timestamp information for the radar data. This absolute timestamp information includes not only the calibrated clock information but also specific flags to indicate that the time is given when the radar and system are in a continuous synchronized state. The time interval for time synchronization between the lidar and the system can be set according to user requirements, typically (but not limited to) 8 seconds or 16 seconds. Simultaneously, a relative timestamp for the radar data is provided based on the original clock (uncalibrated) and the local timer.
[0041] 3. If a synchronization message is received from the time synchronization server on the nth time, but not on the (n+1)th time, then starting from the time the message is not received, the clock information calibrated on the nth time will be continued. Combined with the local timer, an absolute timestamp will be provided for the radar data. This timestamp will contain specific flags to indicate that the time is given when the radar and system are in a state of non-continuous synchronization. Simultaneously, a relative timestamp for the radar data will be provided based on the original clock (uncalibrated) and the local timer.
[0042] Taking a radar located at the end of a local area network (LAN) where only NTP synchronization services are provided as an example, its time synchronization adaptive process is as follows: Figure 1 As shown.
[0043] Adaptive timestamp to measurement time:
[0044] The specific implementation method is as follows:
[0045] The conversion of internal clock information into a timestamp by a lidar system requires a series of hardware and software operations. The shorter the hardware path and the lower the operation latency, the higher the accuracy of the timestamp. (See also...) Figure 2 In the main control circuit of the radar system, the laser emission and time register synchronization module simultaneously transmits the initial reference time T0 to the radar transmission control module and the time register module. The time of laser emission from the radar is denoted as T1, the time of illumination of the target is T2, and the time of receiving the target's echo is T3. t1 = T1 - T0, t2 = T2 - T1, t3 = T3 - T2. There is a fixed delay, t1, between the emission module emitting light and receiving the emission drive command, which can be detected by a specific measuring device. There is a time difference t2 between the emission module emitting light and illuminating the target, and a time difference t3 between the target's echo signal being detected by the receiving module. Since the distance between the target and the radar is variable, the time differences t2 and t3 are variables for targets at different distances, and t2 = t3. The TDC module can accurately calculate the time difference using T1 and T3. Therefore, for each target, t2 can be accurately obtained: t2 = (T3 - T1) / 2. When the data packaging module packages the data, it can assign a precise timestamp to the packaged data based on the reference time T0 and the known time differences t1 and t2. The timestamp time corresponds to T0+t1+t2.
[0046] This method of correcting the timestamp point by point ensures an accurate correspondence between the instant the radar laser illuminates the object being measured and the timestamp, eliminating the time deviation caused by the distance of the object. It also enables adaptive adjustment of the timestamp for targets at different distances.
[0047] An adaptive lidar includes a PHY chip and an FPGA chip;
[0048] Based on the PHY chip, network time protocol servers and network time protocol clients were established for each time synchronization network of the lidar, as well as a TCP / IP module for network communication with external systems.
[0049] Based on the FPGA chip, a crystal oscillator, timing module, and time synchronization module were established for time synchronization; a time calculation module and distance calculation module for data processing; a time register module and data packing module for data and integration; and a light output and time register synchronization module and a transmission control module for light output control.
[0050] To achieve the adaptive functions described above, the hardware modules in the LiDAR perform the following operations:
[0051] 1. Regarding the "Adaptive Network Location and Time Synchronization Protocol" function:
[0052] The FPGA retrieves the crystal oscillator clock information through the timing module. The time synchronization module organizes the clock information into a synchronization message. The NTP / PTP server inside the PHY chip receives the time synchronization request and broadcasts the synchronization message to the NTP / PTP network. The NTP / PTP client inside the PHY chip sends a time synchronization request to the NTP / PTP network, receives the time synchronization message from the NTP / PTP network, and sends the synchronization information in the message back to the timing module through the time synchronization module to achieve timing calibration.
[0053] 2. Regarding the "Adaptive Time Synchronization" function:
[0054] The timing module simultaneously transmits the synchronized absolute timestamp and the relative timestamp corresponding to the original clock to the time register module, which then transmits them to the data packaging module. After the laser echo signal passes through the receiving module and the TDC module to obtain precise time information, it is then processed by the time calculation module and the distance calculation module to obtain radar measurement data. The data packaging module packages the integrated timestamp and radar measurement data and sends it to an external system via the TCP / IP communication module.
[0055] 3. Regarding the "adaptive timestamp to measurement time" function:
[0056] The measurement time calculation module can accurately obtain the laser emission time T1 and the radar reception time T3 of the target echo from the data transmitted by TDC. The time difference t1 between the reference time T0 and the emission time T1 can be obtained through system calibration and directly set as a known fixed value. The measurement time calculation module sends t1 and t2 to the data packaging module. During the data integration and packaging process, t1 and t2 correct the absolute timestamp and relative timestamp to complete the self-adaptation of the timestamp corresponding to the measurement time.
[0057] Where t1 = T1 - T0
[0058] t2=(T3-T1) / 2
[0059] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.
Claims
1. A method for implementing adaptive functionality in a lidar system, characterized in that: The adaptive functions of this lidar include network location adaptation, time synchronization protocol type adaptation, time synchronization function adaptation, and timestamp-corresponding measurement time adaptation. The adaptive implementation method for network location and time synchronization protocol type is as follows: the LiDAR used is applicable to at least two time synchronization networks, and simultaneously includes a network time protocol server and a network time protocol client corresponding to each time synchronization network; when the LiDAR acts as a client, it sends time synchronization request messages for all types of time synchronization networks through its communication port, determines the corresponding time synchronization network type based on the received reply messages, and calibrates the local clock; when the LiDAR acts as a server, it determines the corresponding time synchronization network type based on the received time synchronization request messages, and sends reply messages using the network time protocol server corresponding to that time synchronization network type. The adaptive implementation method of time synchronization function is as follows: the LiDAR used is connected to the time synchronization server through the local area network. The time synchronization server calibrates the LiDAR according to the IEEE1588 standard. The LiDAR adds a relative timestamp generated based on the uncalibrated local clock and an absolute timestamp generated based on the calibrated local clock to the LiDAR data. The absolute timestamp contains a flag bit to indicate whether the time synchronization server is in a continuous synchronization state with the LiDAR. The adaptive implementation method of timestamp corresponding to measurement time is as follows: obtain the initial reference time T0 when the light emission command is sent to the radar transmission control module and the time register module, the light emission time T1 of the lidar, the time when the laser illuminates the target under test T2, and the time when the radar receives the echo of the target under test T3. Obtain the delay duration according to the time interval t1 between T1 and T0 and the time interval t2 between T2 and T1. Then, assign a timestamp to the packaged data generated by the lidar according to T0 and the delay duration. The timestamp is T0+t1+t2, where t1=T1-T0 and t2=(T3-T1) / 2.
2. The adaptive function implementation method of the lidar according to claim 1, characterized in that: The time synchronization networks applicable to lidar include NTP time synchronization networks and PTP time synchronization networks.
3. The adaptive function implementation method of the lidar according to claim 1, characterized in that: The time period for the time synchronization server to synchronize the LiDAR is set according to user requirements.
4. The adaptive function implementation method of the lidar according to claim 1, characterized in that: After the time synchronization server successfully synchronizes the LiDAR, if the LiDAR does not receive a reply message from the time synchronizer in a certain instance, the LiDAR will use the clock information that was calibrated when it last received a reply message from the time synchronizer.
5. A lidar with adaptive function, characterized in that: Including PHY chips and FPGA chips; Based on the PHY chip, network time protocol servers and network time protocol clients were established for each time synchronization network of the lidar, as well as a TCP / IP module for network communication with external systems. Based on the FPGA chip, a crystal oscillator, timing module, and time synchronization module were established for time synchronization; a time calculation module and distance calculation module were established for data processing; a time register module and data packing module were established for data and integration; and a light output and time register synchronization module and a transmission control module were established for light output control. The FPGA retrieves the crystal oscillator clock information through the timing module. The time synchronization module organizes the clock information into a synchronization message. The NTP / PTP server inside the PHY chip receives the time synchronization request and broadcasts the synchronization message to the NTP / PTP network. The NTP / PTP client inside the PHY chip sends a time synchronization request to the NTP / PTP network, receives the time synchronization message from the NTP / PTP network, and sends the synchronization information in the message back to the timing module through the time synchronization module to achieve timing calibration, thus realizing the self-adaptation of network location and time synchronization protocol type. The timing module transmits the synchronized absolute timestamp and the relative timestamp corresponding to the original clock to the time register module, and then to the data packaging module. After the laser echo signal passes through the receiving module and the TDC module to obtain accurate time information, it passes through the time calculation module and the distance calculation module to obtain radar measurement data. The data packaging module packages the integrated timestamp and radar measurement data and sends them to the external system through the TCP / IP communication module to achieve adaptive time synchronization. The measurement time calculation module obtains the laser emission time T1 and the radar echo time T3 from the data transmitted by the TDC. The time difference t1 between the reference time T0 and the emission time T1 is obtained through system calibration. The measurement time calculation module sends t1 and t2 to the data packaging module. During the data integration and packaging process, t1 and t2 are corrected for the absolute timestamp and the relative timestamp to achieve self-adaptation of the timestamp corresponding to the measurement time. Where t1 = T1 - T0, t2 = (T3 - T1) / 2.