EM4

1 Safety Notices
1.1 Legal Statement
Warning
Unless otherwise stated, all rights (including copyrights, trademarks, patents, trade secrets, and other related rights) in RoboSense's products, technologies, software, programs, data, and other information (including text, icons, photographs, audio, video, graphics, color combinations, layout design, etc.) are owned by RoboSense and its licensors.
No one may use any content contained in this manual in any unauthorized manner without the prior written consent of RoboSense.
The word "RoboSense" and other logos and product and service names are owned by RoboSense. If you need to use them for any advertising or displaying purposes, you must obtain prior written authorization from RoboSense.
1.2 User Guidelines
Please use this product in accordance with the following requirements
- Please strictly abide by relevant national laser safety laws and regulations;
- Please read this product manual in detail before using the product;
- Please use this product only in the relevant field of application;
- Please avoid using this product in environments that are explosive, highly corrosive, or beyond the IP protection level of the equipment.
1.3 Illegal Operation
Please use this product in accordance with the regulations, otherwise it may cause product damage, property loss, and personal injury. Users are responsible for risk arising from unauthorized operations.
- Do not disassemble or modify this product (including accompanying accessories);
- Non-specified power supply and accompanying accessories are prohibited;
- Please avoid abnormal operations such as dropping, colliding, burning, etc.;
- If you notice any damage to the appearance of the device (arc protection cover), please immediately stop using it;
- If you notice any abnormal operation of the product, please immediately stop using it and contact RoboSense in a timely manner.
1.4 Requirements for Operating Personnel
The use of this product requires certain basic professional knowledge and other related requirements for operating personnel. Unreasonable operations performed by personnel without basic knowledge or training do not constitute a fault of RoboSense and may cause damage to equipment and personal property.
- Please read the product manual in detail before using the device;
- Prohibit illegal operations;
- Before working, personnel must undergo training and obtain relevant construction qualifications;
- Have some basic knowledge of computer data connection, electrical, and so on.
1.5 Work Safety and Special Hazards
To avoid risks of accidents, damage to sensor or violating of your product warranty, please read and follow the instructions in this manual carefully before operating the product.
- Laser Safety: This product meets the following standards for laser products: IEC 60825-1:2014;

- High Temperature Warning: Please pay attention to the overheating sign on the LiDAR surface to avoid a hot LiDAR surface that may lead to sensor failure or undesirable consequences;

-
Retain Instructions: The safety and operating instructions should be retained for future reference;
-
Heed Warnings: All warnings on the product and in the operating instructions should be adhered to;
-
Servicing: Except for what's described in this manual, the sensor has no field serviceable parts. For servicing, please contact RoboSense sales or the authorized distributors.
2 Product Description
The following content describes the status and functions of the EM4-F B prototype. After the release of the new version of the prototype, the product manual will be updated to the latest status synchronously.
2.1 Product Overview
The EM4-F is a high-performance automotive-grade LiDAR based on VCSEL + SPAD-SoC digital transceiver technology and a one-dimensional rotating mirror scanning system. It features 13 vertically arranged emission blocks, with each zone corresponding to 40-line reception via SPAD-SoC, totaling 520 lines, as shown in Figure 1. The EM4 has a maximum detection range of 300 meters, offering a resolution of \(0.1^{\circ}\) (H) \(\times\) \(0.05^{\circ}\) (V) in non-ROI areas and achieving up to \(0.05^{\circ}\) (H) \(\times\) \(0.05^{\circ}\) (V) in ROI regions.

Figure 1 Operating Principle Diagram of EM4-F
2.2 Product Structure
The structure of EM4-F LiDAR is illustrated in Figure 2.

Figure 2 Product Structure of EM4-F LiDAR (Optical center refers to polygon mirror's rotational center)
2.3 FOV Distribution
EM4-F LiDAR's FOV is showed in Table 1 and Figure 3.
Table 1 EM4-F FOV
| Specifications | Horizontal FOV | Vertical FOV |
|---|---|---|
| 120° × 25° FOV | -60° ~ +60° | -12.5° ~ +12.5° |

Figure 3 EM4-F LiDAR FOV
2.4 Specifications
Table 2 Parameter specifications
| Sensor | |||
|---|---|---|---|
| Measurement Range1 | ≥250m @10% NIST, 100klux | Accuracy2 (Typical) | ±5 cm@1σ |
| Horizontal FOV | 120° (-60° ~ +60°) ROI: 40° (-20° ~ +20°) |
Horizontal Angle Resolution | 0.1° ROI: 0.05° |
| Vertical FOV | 25° (-12.5° ~ +12.5°) | Vertical Angle Resolution | 0.05° |
| Output | |||
| PPS |
NROI MODE: Single Return: 6,240,000 pts/s Dual Return: 12,480,000 pts/s ROI MODE: Single Return: 8,320,000 pts/s Dual Return: 16,640,000 pts/s * Dual Return in ROI mode is supported only in compress mode. |
||
| Ethernet Output | 1000Base-T1 | ||
| Communication Protocols | UDP | ||
| Data Package | 3D spatial coordinates, reflection intensity, timestamps, etc. | ||
| Mechanical and Electrical Specifications | |||
| Supply Voltage | 12V (9V~16VDC) | Main Body Dimension | Depth 150mm * Width 120mm * Height 45mm * Full-scale dimensions are shown in Figure 1 |
| Power Consumption3 | 15W (Typical) | Operating Temperature4 | -40°C ~ +85°C |
| Weight | <1000g | Storage Temperature | -40°C ~ +105°C |
| Time Synchronization | gPTP | Protection Level | IP67/IP6K9K |
| Frame | 10 Hz | Blind Zone | 1m |
1 The ranging capability is based on a 10% NIST diffuse reflector as the target, under 100 KLux ambient illuminance at 90% probability of detection (PoD);
2 The ranging accuracy is based on a 90% NIST diffuse reflector as the target, and the test results may be affected by environmental factors, including but not limited to environmental temperature, target object distance, etc;
3 Power consumption test results may be affected by external environmental factors, including but not limited to ambient temperature, target distance, and target reflectivity.
4 Operating temperature may vary depending on external conditions, including but not limited to lighting conditions and airflow changes.
2.5 Time Synchronization Mode
EM4-F currently supports gPTP (IEEE802.1AS protocol) synchronization mode. If users have special requirements, please contact RoboSense.
2.5.1 gPTP synchronization principle
gPTP (General Precise Time Protocol, IEEE802.1AS protocol) is a derivative protocol of PTP in Time Sensitive Networking. The synchronization mechanism adopts the Peer Delay Mechanism, which is consistent with the PTP protocol, and also uses Ethernet L2 layer communication. Unlike PTP, gPTP requires the use of hardware timestamp, which means hardware timestamp, so the requirements for switches and Master clocks are more stringent and must meet the IEEE802.1AS protocol.
2.5.2 gPTP wiring method
To use gPTP synchronization method, the following preparations need to be made:
- GPTP Master timing host (plug and play, no additional configuration required);
- Ethernet switch;
- Devices that support gPTP protocol for waiting to be taught.
Note
- The Master timing device is a third-party device and is not included in RoboSense shipments. Users need to purchase it themselves.
- The RoboSense product, as a Slave device, only obtains the time sent by the Master and does not judge the accuracy of the Master clock source. If there is a sudden change in the time when analyzing the LiDAR point cloud, please check if the time provided by the Master is accurate.
- After the synchronization of the LiDAR, the Master disconnects, and the time in the point cloud data packet will be superimposed according to the internal clock of the LiDAR. The LiDAR will only be reset after it is powered off and restarted.
2.6 Frame Synchronization Method
The EM4-F's SoC transmits synchronization pulses to the scanning mirror every 400ms. In the default configuration, at each whole 100ms time point, the mirror is positioned at the starting point of the FOV scan*. The LiDAR scans from left to right, with the mirror coordinate system's \(15^{\circ}\) position (corresponding to \(-60^{\circ}\) in FOV, which is the starting position of FOV scan) as shown in Figure 4.
Starting from software version V07.00.0A, support is added for "the mirror to be at the starting position of FOV scanning at whole hundred millisecond time points." Previous software versions allowed adjustment of frame scanning start time, but under default settings, the mirror would not be at the FOV scanning starting position at whole hundred millisecond time points.

Figure 4 Mirror Coordinate System and FOV Coordinate System
Software versions prior to V07.00.0A supported adjusting the FOV scan start time by configuring registers, with a step size of 1 μs. Refer to Table 3 for related register information.
Table 3 Frame Synchronization Register Information
| Register Address | Register Length | Operation | Value Range | Unit |
|---|---|---|---|---|
| 0x83C4011C | 4 Bytes | Read/Write | 0~1000000 | us |
Register configuration can be implemented using the LiDARAssistant tool. For details, refer to the LiDARAssistant tool documentation.
Software versions V07.00.0A and later will use DID commands to adjust the FOV scan start time.
3 Product Installation
3.1 LiDAR Wiring and Interface Instructions
3.1.1 Interface and definition of onboard Ethernet harness
EM4-F uses one automotive Ethernet and power supply two in one connector, and the matching wiring harness is shown in Figure 5.

Figure 5 Vehicle Ethernet power harness
The definitions of EM4-F vehicle Ethernet power harness connectors and pins are shown in Table 4
Table 4 Definition of Ethernet power harness interface for onboard vehicles
| A side | Definition | Description | B side | |
|---|---|---|---|---|
| J1 (TE 2397380-1) |
1 | POWER | Power supply | J3 (Molex 43025-0600) |
| 2 | Wakeup | Wake signal | ||
| 3 | / | / | ||
| 4 | GND | Ground | ||
| 5 | / | / | ||
| 6 | / | / | ||
| 7 | 1000BASE-T1 N | Automotive Ethernet differential pair | J2 (Amphenol NTHCF011A10S) |
|
| 8 | 1000BASE-T1 P | Automotive Ethernet differential pair | ||
3.1.2 Interface Box

Figure 6 Interface Box
The EM4-F interface box has a power indicator light and various interfaces, as shown in Figure 6, which can be connected to power input and RJ45 network port.
Table 5 Definition of Interface Box
| Interface | Interface Name | Function Description |
|---|---|---|
| J1 | Power Wake-up Signal | Provide power to the radar and output wake-up signals |
| J2 | Network | 1000BASE-T1 vehicle Ethernet interface |
| J3 | DC Power connector | External power input |
| J4 | Switch | Wake up signal control switch, when the switch is pressed, the wake-up signal is turned on |
| J5 | RJ45 | 1000BASE-TX Industrial Ethernet |
3.1.3 Power interface
The EM4-F interface box power supply uses a standard DC 5.5-2.1 interface.
When the power input is normal, the green indicator light on the power box remains on. When the green indicator light goes out, please check if the power input is normal. If the power input is normal, the interface box may be damaged. Please contact RoboSense.
3.1.4 RJ45 Ethernet port
The interface box only supports Gigabit Ethernet, and when using the interface box, the network interface uses a standard RJ45 interface.
3.2 Quick Connection
The EM4-F network parameters can be configured, and the factory default uses a fixed IP and port number mode. Please refer to Table 6 for details.
Table 6 Factory default network configuration table
| Equipment | IP address | MSOP package port number |
DIFOP1 packet port number |
DIFOP2 packet port number |
|---|---|---|---|---|
| EM4-F | 192.168.1.200 | 6699 | 7788 | 7766 |
| PC | 192.168.1.102 |
Note
- Main Data Stream Output Protocol (MSOP) encapsulates scanned LiDAR data including distance, yaw angle, reflectivity and other parameters into output packets.
- Device Information Output Protocol (DIFOP1) encapsulates LiDAR status information into output packets.
- Device Information Output Protocol (DIFOP2) encapsulates pitch angle data for each SPAD-SoC channel (1-520), yaw angle compensation values for different VCSEL zones, and pitch angle compensation values for each mirror facet into output packets.
When users use the product, they need to set the computer's IP to be on the same network segment as the product, such as 192.168.1.x (x values range from 1 to 254), with a subnet mask of 255.255.255.0.
Unknown product network configuration information, please connect the product and use Wireshark to capture the product output packet for analysis. Configure the IP and connection method as follows.
-
Connect the LiDAR
The connection method is shown in Figure 7.
- The LiDAR is connected to the interface box through a vehicle mounted Ethernet power harness;
- The PC and interface box are connected using Gigabit Industrial Ethernet through RJ45 network port connectors;
- After power on, under normal working conditions, the green indicator light of the interface box of the LiDAR will remain on, and the location details of the indicator light are shown in Figure 7.
Figure 7 Schematic diagram of interface box connection
-
Capture packets through Wireshark, parse ARP packets for local IP configuration
- As mentioned above, after the laser radar is connected to the PC, start Wireshark (a third-party network analysis tool), select the correct network port, and start capturing packets;
- Enter "ARP" in the search box of Wireshark to search for mutual addressing messages between the LiDAR and PC, as shown in Figure 8;
Figure 8 Analyzing ARP Messages
- As shown in Figure 8, the word SutengIn in the Source column indicates the information source of the LiDAR, and prompts 192.168.1.200 as the Source IP, which is the LiDAR IP. Then request access to 192.168.1.102, which is the PC IP. If the local IP is not the requested IP, the local IP of the PC needs to be configured as 192.168.1.102. For details, please refer to step 3; If it can be accessed normally, proceed to step 4.
-
Configure the local IP of the PC
- In the control panel, enter the "Network and Sharing Center" through "Network and Internet", click on the corresponding Ethernet connection in the "View Active Networks" content, enter the corresponding "Ethernet Status", and click on the "Properties" settings;
- Double click on Internet Protocol version 4 (TCP/IPv4) to enter IP information settings and configure using a static IP address;
- Set the local IP address to 192.168.1.102 and subnet mask to 255.255.255.0. Click "Confirm" to complete the static IP setting for the PC.
-
Connection completed.
Note
- The time synchronization module (gPTP) is not a factory standard product. If you need to use related functions, please purchase them yourself.
- The above configuration of local static IP is only for Windows system operation as an example. For other operating systems, please refer to the actual situation.
- EM4-F adopts a static ARP list, which only sends ARP packets after the radar is powered on and before it is connected to the upper computer. If the upper computer is replaced after normal communication between the radar and the upper computer, the radar needs to be powered on again to communicate with the new upper computer.
4 Product usage
4.1 Product Coordinate System
The vehicle coordinate system is defined as follows:
- Coordinate origin: center of the front axle
- \(X\)-axis: positive forward
- \(Y\)-axis: positive to the left
- \(Z\)-axis: positive vertically upward
As shown in Figure 9:

Figure 9 Schematic diagram of vehicle coordinates
The product's coordinate system origin and direction are defined as follows:
- Coordinate origin: LiDAR optical center
- \(X\) axis: forward is positive
- \(Y\) axis: left is positive
- \(Z\) axis: vertically upward is positive
As shown in Figure 10:

Figure 10 Schematic diagram of LiDAR coordinates
4.2 RView Usage
In terms of data visualization for EM4-F, free tools such as Wireshark and tcpdump can be used to capture raw data, while RSView can help users visualize raw data more conveniently.
4.2.1 Software Functions
RSView provides real-time visualization of EM4-F data. Its core features and supported formats are as follows:
- Data recording and playback: Supports recording real-time data and saving it in
.pcapfile format. The software currently does not support.pcapngformat files. - Multi-dimensional data display: Displays the measured distance values as points (point cloud). Supports multiple custom colors to display data, such as: reflection intensity, time, distance, horizontal angle, and laser beam index.
- Data export: The displayed data can be exported and saved in
.csvformat.
RSView includes the following features:
- Real time display of data through Ethernet;
- Save real-time data records as PCAP files;
- Replay from the recorded PCAP file;
- Different types of visualization modes, such as distance, time, horizontal angle, etc;
- Display point data in a table;
- Export point cloud data to CSV format file;
- Measuring distance tools;
- Display multiple consecutive frames of playback data simultaneously;
- Show or hide individual wiring harnesses (channels) in EM4-F;
- Crop display.
4.2.2 Installing RSView
- Supported systems: Windows 64-bit, Ubuntu 18.04 and above.
- Download link: Download the latest version of the RSView software installation package
Important Installation Notes
- Path restriction: Please do not include Chinese characters in the software extraction path.
- No installation required: This software is a portable version that does not require installation. After extraction, simply run the executable file to use it normally.
4.2.3 Using RSView
After connecting the LiDAR to the PC, open RSView and follow the steps below to stream point cloud data online.
First, select File → Open Sensor from the top-left menu bar.

In the pop-up window, choose RSEM4 as the Sensor Type, then click OK.

In the newly opened window, configure the LiDAR protocol port numbers. The default configuration is as follows. After confirming, click OK to play the point cloud online:
- MSOP Port (MSOP packet port number): default is
6699 - DIFOP Port (DIFOP2 packet port number): default is
7766

You can open the software user guide by pressing the F1 key, or access it through the software menu bar via Help → RS-LiDAR User Guide.
4.3 Communication Protocol
EM4-F communicates with the computer using Ethernet as the physical medium, with the upper transport layer using the UDP protocol.
Its output data packets are mainly divided into the following three core types:
-
MSOP packets
-
Fixed format: the fixed length is
1084 bytes -
Compressed format: packet length is not fixed, dynamically split into packets
-
-
DIFOP1 packets
The fixed length of the protocol packets is
256 bytes -
DIFOP2 packets
The fixed length of the protocol packets is
1162 bytes
For the compressed format of MSOP, the data packet follows the following packet-splitting mechanism during network transmission:
- When the total length of the data packet is greater than
1396 Bytes, packet splitting starts automatically - The length of the first split packet is fixed at
1320 Bytes
The communication between the EM4-F and the computer utilizes Ethernet medium with UDP protocol, and the communication protocols are categorized into three types as detailed in Table 7.
Table 7 List of Product Protocol
| (Protocol/Package) Name | Abbreviation | Function | Type | Package size | Sending interval |
|---|---|---|---|---|---|
| Main data Stream Output Protocol | MSOP | Scan data output | UDP | Fixed Mode: 1084 bytes | NROI + Single Return: 24000 times/s |
| NROI + Dual Return: 48000 times/s | |||||
| ROI + Single Return: 32000 times/s | |||||
|
Compress Mode: 1. Non-fragmented packets: variable length, maximum 1396Bytes 2. fragmented packets: first packet 1320Bytes, second packet: variable length |
/ | ||||
| Device Information Output Protocol 1 | DIFOP1 | Product information output | UDP | 256 Bytes | DIFOP1: 100 times |
| Device Information Output Protocol 2 | DIFOP2 | Product information output | UDP | 1162 Bytes | DIFOP2: 0.5 time |
4.3.1 Main data Stream Output Protocol (MSOP)
I/O type: Product output, computer parsing.
The default port number at the factory is 6699.
The fundamental structure in Fixed Mode is illustrated in Figure 11 below:

Figure 11 MSOP Packet Data Packet Definition Schematic Diagram
For detailed definitions, see Table 8.
Table 8 Datasheet of MSOP
| Content | Offset (Byte) | Length (Byte) | Instruction |
|---|---|---|---|
| pkt_head | 0 | 4 | header: 55aa5aa5 |
| pkt_cnt | 4 | 2 |
1 Non Compress MODE 1.1 NROI MODE: Single Return: 1~2400 *Each frame point cloud contains 1200 columns, each column is divided into 2 packets, with 260 pixels packaged per packet. The first and second packets correspond to rows 1-260 and 261-520 of the column respectively, and so on. Dual Return: 1~4800 *Each frame point cloud contains 1200 columns, each column is divided into 4 packets, with 130 pixels packaged per packet. The first to fourth packets correspond to rows 1-130, 131-260, 261-390 and 391-520 of the column respectively, and so on. 1.2 ROI MODE: Single Return: 1~3200 *Each frame point cloud contains 1600 columns, each column is divided into 2 packets, with 260 pixels packaged per packet. The first and second packets correspond to rows 1-260 and 261-520 of the column respectively, and so on. Dual Return: not supported |
| ROI_compression_mode | 6 | 1 | 0x00: without compress + NROI 0x01: without compress + ROI |
| reserved | 7 | 1 | Reserved |
| return_mode | 8 | 1 | 0: Dual Return 4: Strongest Return 5: Last Return 6: Nearest Return |
| time_mode | 9 | 1 | Time synchronization method: 0x00: Internal time 0x03: gPTP |
| timestamp | 10 | 10 | Timestamp Byte0~5: s Byte6~9: us |
| frame_sync | 20 | 1 |
Frame synchronization 0x00: Not synchronized 0x01: Synchronized Synchronization: The FOV scan initiation timing exhibits less than 0.6ms jitter while maintaining synchronization stability for over 50ms. |
| frame rate | 21 | 1 | 0x0A: 10Hz |
| column_num | 22 | 2 | NROI Column Number: 0~1199 ROI Column Number: 0~1599 |
| yaw_angle | 24 | 2 | yaw angle: factor: 0.01 deg |
| pack mode | 26 | 1 | Compress/Non Compress MODE 0x01: Non Compress MODE |
| surface id | 27 | 1 | Surface 0x01: A 0x02: B 0x03: C 0x04: D |
| vcsel_internal | 28 | 1 | Timing Offset Between Odd/Even VCSEL Zones Signed number, range: -128~127us; central value: 256us |
| reserved | 29 | 1 | reserved |
| lidar_type | 30 | 1 | Default: 0x70 |
| main_temp | 31 | 1 |
Main FPGA temperature offset: -80 Unit: °C * Real Temperature = (value-80)°C |
| point data 1-260 | 32 | 1040 |
260-pixel Point Cloud Data 1) Single Return Mode: For packet 2m-1 (m=1,2,...,1200/1600), point data n corresponds to the echo from the nth pixel; for packet 2m (m=1,2,...,1200/1600), point data n corresponds to the echo from the (n+260)th pixel. 2) Dual Return Mode: For packet 4m-3 (m=1,2,...,1200), point data 2n-1 (n=1,2,...,130) represents the first echo from the nth pixel while point data 2n (n=1,2,...,130) represents the second echo from the nth pixel; for packet 4m-2 (m=1,2,...,1200), point data 2n-1 corresponds to the first echo from the (n+130)th pixel and point data 2n corresponds to the second echo from the (n+130)th pixel; for packet 4m-1 (m=1,2,...,1200), point data 2n-1 indicates the first echo from the (n+260)th pixel and point data 2n indicates the second echo from the (n+260)th pixel; for packet 4m (m=1,2,...,1200), point data 2n-1 reflects the first echo from the (n+390)th pixel and point data 2n reflects the second echo from the (n+390)th pixel. |
| DataLength | 1072 | 2 | E2E Profile4 Data Length: 04 3C |
| Counter | 1074 | 2 | E2E Profile4 Counter: 00 00~FF FF |
| DataId | 1076 | 4 | E2E Profile4 Data Id: 00 00 0E 5D |
| Crc32 | 1080 | 4 | E2E Profile4 Crc32 |
The detailed definition of each point data can be found in Table 9:
Table 9 Packet definition of point data n
| point data (4 Bytes) | |||
|---|---|---|---|
| content | offset(byte) | byte | instruction |
| radius | 0 | 2 | In polar coordinate system, the radial point distance value has a distance resolution of 5mm |
| intensity | 2 | 1 | reflectivity range: 0~255 |
| point_attribute | 3 | 1 | point attribute 0x00: normal point 0x08: noise point |
In compress mode, the detailed definitions of the first and second MSOP packets are specified in Table 10 and Table 11 respectively:
Table 10 First MSOP Packet Definition
| Content | Offset (Byte) | Length (Byte) | Instruction |
|---|---|---|---|
| pkt_head | 0 | 4 | header: 55aa5aa5 |
| Pkt_cnt | 4 | 2 |
Packet count indication. In compressed mode, the total number of packets per frame is variable. For single return mode: The point cloud data of each column is compressed. If the compressed packet length is ≤1396 bytes, one packet corresponds to one column; if the packet length exceeds 1396 bytes, one column corresponds to two packets. The total packet count is variable. For dual return mode: The first echo of each column is compressed first. If the compressed data length of the first echo is ≤1396 bytes, it occupies one packet; if exceeding 1396 bytes, it occupies two packets. Then, the second echo is compressed: if the compressed data length is ≤1396 bytes, it occupies one packet; if exceeding 1396 bytes, it occupies two packets. |
| ROI_compression_mode | 6 | 1 | 0x02: compress + NROI 0x03: compress + ROI |
| reserved | 7 | 1 | Reserved |
| return_mode | 8 | 1 | 0: Dual Return 4: Strongest Return 5: Last Return 6: Nearest Return |
| time_mode | 9 | 1 | Time synchronization method: 0x00: Internal time 0x03: gPTP |
| timestamp | 10 | 10 | Timestamp Byte0~5: s Byte6~9: us |
| frame_sync | 20 | 1 |
Frame synchronization 0x00: Not synchronized 0x01: Synchronized Synchronization: The FOV scan initiation timing exhibits less than 0.6ms jitter while maintaining synchronization stability for over 50ms. |
| frame rate | 21 | 1 | 0x0A: 10Hz |
| column_num | 22 | 2 | NROI Column Number: 0~1199 ROI Column Number: 0~1599 |
| yaw_angle | 24 | 2 | yaw angle: factor: 0.01 deg |
| pack mode | 26 | 1 |
bit0-3: Compress mode flag 3: compressed mode for compress mode, packets exceeding 1396 bytes are split with the first split packet fixed at 1320 bytes. bit7-4: Packet fragmentation flag 0: unsplit packet 1: first fragment 2: second fragment |
| surface id | 27 | 1 | Surface 0x01: A 0x02: B 0x03: C 0x04: D |
| vcsel_internal | 28 | 1 | Timing Offset Between Odd/Even VCSEL Zones Signed number, range: -128~127us; central value: 256us |
| reserved | 29 | 1 | reserved |
| lidar_type | 30 | 1 | Default: 0x70 |
| main_temp | 31 | 1 |
Main FPGA temperature offset: -80, Unit: °C * Real Temperature = (value-80)°C |
| point data | 32 | x(x≤1363) |
The radial distance (2 bytes) of all pixels in a full column for a single echo forms the first compression block. The reflectivity (intensity) and point attributes (2 bytes) form the second compression block. If the length of the first MSOP packet exceeds 1396 bytes, it is split into two packets. The first packet has a length of 1320 bytes. |
Table 11 Second MSOP Packet Definition
| content | offset(Byte) | Length(Byte) | Instruction |
|---|---|---|---|
| pkt_head | 0 | 4 | header: 55aa5a02 |
| pkt_cnt | 4 | 2 |
Packet count indication. In compressed mode, the total number of packets per frame is not fixed. If the length of the first MSOP packet exceeds 1396 bytes, it will be split into two packets, in which case a second MSOP packet will exist. |
| reserved | 6 | 2 | reserved |
| point data | 8 | y |
The radial distance (2 bytes) of all pixels in an entire column for a single echo forms the first compression block. The reflectivity (intensity) and point attributes (2 bytes) form the second compression block. If the length of the first MSOP packet exceeds 1396 bytes, it is split into two packets. The first packet has a fixed length of 1320 bytes, while the length of the second packet is variable. |
| DataLength | 8+y | 2 | E2E Profile4 Data Length: 04 3C |
| Counter | 10+y | 2 | E2E Profile4 Counter: 00 00~FF FF |
| DataId | 12+y | 4 | E2E Profile4 Data Id: 00 00 0E 5D |
| Crc32 | 16+y | 4 | E2E Profile4 Crc32 |
The structure of the compressed point data blocks is shown in Table 12.
Table 12 Compressed Block Structure
| Mode | Total Length | Initial Value | Start Forward Difference | ... | End Forward Difference | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 2bit | 14bit | 16bit | 4bit | 4bit | 4bit | 4bit | ... | 4bit | 4bit | 4bit | 4bit |
| 0 | 1 | Initial Value | - | ... | - | ||||||
| 1 | n+1 | Initial Value | diff_0 | diff_1 | diff_2 | diff_3 | ... | diff_n-4 | Padding 0 | Padding 0 | Padding 0 |
| 2 | n+1 | Initial Value | diff_0 | diff_1 | ... | diff_n-2 | Padding 0 | ||||
| 3 | n+1 | Initial Value | org_0 | ... | org_n-1 | ||||||
4.3.2 Device Information Output Protocol (DIFOP1&DIFOP2)
I/O type: Product output, computer parsing.
EM4-F outputs two different types of DIFOP data packets during operation.
| Protocol Type | Factory Default Port Number | Transmission Period |
|---|---|---|
| DIFOP1 | 7788 |
10ms |
| DIFOP2 | 7766 |
2s |
Users can interpret the specific information of various parameters of the currently used device by reading DIFOP1.
The DIFOP1 data packet mainly encapsulates the following core device parameters:
- Basic device information: device serial number (\(S/N\)), firmware version information.
- Compatibility and network: host driver compatibility information, network configuration information.
- Operation and diagnostics: calibration information, motor operation configuration, operating status, fault diagnosis information.
Users can parse the point cloud by reading DIFOP2.
The DIFOP2 packet mainly encapsulates the following core compensation parameters:
- Pitch angle information: The angle (\(pitch\)) information corresponding to each line (\(1 \sim 520\)) of the LiDAR SPAD-SoC.
- Yaw angle compensation: Compensation information for the angles (\(yaw\)) of different VCSEL partitions.
- Rotating mirror compensation: Compensation information for the angle (\(pitch\)) of each face of the rotating mirror.
The basic structure of the data packet is shown in Table 13,14.
Table 13 DIFOP1 Data Sheet
| DIFOP1 | Content | Offset | Length | Instruction |
|---|---|---|---|---|
| Header | StatusHdr | 0 | 4 | A5 FF 00 5A |
| Version | Reserved | 4 | 20 | |
| SW Version | 24 | 3 | ||
| HW Version | 27 | 2 | ||
| SerialNumber | IntSN | 29 | 6 | |
| CusSN | 35 | 16 | Customer SN | |
| WorkInformation | Reserved | 51 | 1 | |
| FrameRate | 52 | 1 | 0x0A: 10Hz | |
| WaveMode | 53 | 1 | 0x00: DualReturn 0x04: StrongestReturn |
|
| Reserved | 54 | 10 | ||
| Lidar_Heater_Status | 64 | 1 | bit0: Lidar_Heater_Switch 0b0: Heating off 0b1: Heating on bit1-7: Reserved |
|
| Reserved | 65 | 24 | ||
| TimeSyncInformation | TimesyncMode | 89 | 1 | 0x00: internal local timer 0x03: gPTP timer |
| TimesyncStatus | 90 | 1 | 0x00: failed 0x01: Success 0x02: Timeout |
|
| TimeStamp | 91 | 10 | 0-5 bytes: Second 6-9 bytes: MicroSecond |
|
| NetParameter | PhyMasterSlaveMode | 101 | 1 | 0x02: slave |
| SrcIP | 102 | 4 | 192.168.1.200 | |
| NetMask | 106 | 4 | 255.255.255.0 | |
| MacAddress | 110 | 6 | ||
| MsopDstIp | 116 | 4 | 192.168.1.102 | |
| MsopSrcPort | 120 | 2 | 6699 | |
| MsopDstPort | 122 | 2 | 6699 | |
| Difop1DstIp | 124 | 4 | 192.168.1.102 | |
| Difop1SrcPort | 128 | 2 | 7788 | |
| Difop1DstPort | 130 | 2 | 7788 | |
| Difop2DstIp | 132 | 4 | 192.168.1.102 | |
| Difop2SrcPort | 136 | 2 | 7766 | |
| Difop2DstPort | 138 | 2 | 7766 | |
| DoIPDstIp | 140 | 4 | 192.168.1.102 | |
| DoIPSrcPort | 144 | 2 | 13400 | |
| Reserved | 146 | 10 | ||
| Voltage & Temp | MCU_VMON_RX_D1V1 | 156 | 2 | |
| MCU_VMON_F_1V0 | 158 | 2 | ||
| MCU_VMON_F_1V8 | 160 | 2 | ||
| MCU_VMON_F_2V5 | 162 | 2 | ||
| MCU_VMON_M_3V3 | 164 | 2 | ||
| MCU_VMON_A_3V3 | 166 | 2 | ||
| MCU_VMON_WAKE_EXT | 168 | 2 | ||
| MCU_IMON_WINDOW | 170 | 2 | ||
| MCU_VMON_WINDOW | 172 | 2 | ||
| MCU_VMOM_SYS_5V | 174 | 2 | ||
| MCU_VMOM_VIN | 176 | 2 | ||
| PL_VMOM_M_1V2 | 178 | 2 | ||
| PL_VMON_CHG | 180 | 2 | ||
| PL_VMON_VOP | 182 | 2 | ||
| RX_VT4_N | 184 | 2 | ||
| RX_3V3 | 186 | 2 | ||
| Res3 | 188 | 4 | ||
| TEMP_RX_Sensor | 192 | 1 | Phy = INT-100, °C | |
| TEMP_FPGA1 | 193 | 1 | Phy = INT-100, °C | |
| TEMP_MCU | 194 | 1 | Phy = INT-100, °C | |
| TEMP_MOTOR | 195 | 1 | Phy = INT-100, °C | |
| TEMP_FPGA2 | 196 | 1 | Phy = INT-100, °C | |
| TEMP_TXR1 | 197 | 1 | Phy = INT-100, °C | |
| TEMP_RX | 198 | 1 | Phy = INT-100, °C | |
| TEMP_WINDOW | 199 | 1 | Phy = INT-100, °C | |
| TEMP_TXR2 | 200 | 1 | Phy = INT-100, °C | |
| Reserved | 201 | 5 | ||
| Humidity_Sensor_Value | 206 | 1 | Phy = INT* 1, % | |
| Temperature_Sensor_Value | 207 | 1 | Phy = INT-100, °C | |
| Dew_Point | 208 | 1 | Phy = INT-100, °C | |
| Reserved | 209 | 7 | ||
| Fault | Internal_Power_Supply_Fault | 216 | 3 | |
| LiDAR_Temperature_Fault | 219 | 3 | ||
| Internal_Software_Fault | 222 | 3 | ||
| Internal_Performance_Fault | 225 | 4 | ||
| LidarFunctionFault | 229 | 1 | bit0: Window_Blockage_Error 0b0: False 0b1: True bit1: gPTP_Sync_Error 0b0: False 0b1: True bit2-7: Reserved |
|
| ExtPowerSupplyFault | 230 | 1 | bit0: Battery_High 0b0: False 0b1: True bit1: Battery_Low 0b0: False 0b1: True bit2-7: Reserved |
|
| External_Communication_Fault | 231 | 2 | ||
| FaultStatus | Reserved | 233 | 11 | |
| E2E | DataLength | 244 | 2 | 0x0100 |
| Counter | 246 | 2 | 0x0000-0xFFFF | |
| DataId | 248 | 4 | 0x00000E5C | |
| Crc32 | 252 | 4 |
Table 14 DIFOP2 Data Sheet
| Difop2 | Content | Data Offset | Length | Instruction |
|---|---|---|---|---|
| Header | InfoHdr | 0 | 4 | A5 FF 00 AE |
| Reserved | Res0 | 4 | 63 | |
| AngleInformation | SurfaceCnt | 67 | 1 | 0x04 |
| HalfVcselPixelCnt | 68 | 1 | 0x14 | |
| HalfVcselCnt | 69 | 1 | 0x1A | |
| HalfVcselYawOffset | 70 | 26 | byte0-1: vcsel 1 ... byte24-25: vcsel 13 |
|
| PixelPitch[1~520] | 96 | 1040 | byte0-1: pixel 1 pitch ... byte1038-1039: pixel 520 pitch |
|
| SurfacePitchOffset | 1136 | 8 | byte0~1: Surface A Pitch Offset byte2~3: Surface B Pitch Offset byte4~5: Surface C Pitch Offset byte6~7: Surface D Pitch Offset |
|
| Reserved | 1144 | 6 | ||
| E2E | DataLength | 1150 | 2 | |
| Counter | 1152 | 2 | 00 00~FF FF | |
| DataId | 1154 | 4 | 00 00 5A A5 | |
| Crc32 | 1158 | 4 |
4.4 Point Cloud Data Parsing Guide
EM4-F obtains timestamp information by parsing the timestamp field in the MSOP packet, with a total length of 10 bytes:
The first 6 bytes are s, representing the number of seconds elapsed since 1970-01-01 00:00:00 (UTC time).
The last 4 bytes are us, storing the number of microseconds (\(0 \sim 999999\)).
After reading the time, first convert the hexadecimal number into a decimal second count, use 1970-01-01 00:00:00 as the base to calculate the UTC time, and then add the time difference of the local time zone to obtain the actual time.
The odd and even partitions of the same VCSEL column have different times. The timestamp indicates the time when detection of the entire column is completed, that is, the time of the even-partition pixels. EM4-F obtains the time difference between the odd and even VCSEL partitions by parsing the vcsel_interval field in the MSOP packet. The format is a 1-byte signed number (two's complement), with a range of \(-128 \sim 127\mu\text{s}\) and a center value of \(256\mu\text{s}\). Subtracting the time difference from the even-partition time gives the odd-partition time.
An example
Known conditions:
-
The read value of the
timestampfield in anMSOPpacket is:00 00 00 68 7E 7F 12 00 00 00 64 -
The read value of the
vcsel_intervalfield is:01 00 00 11
Calculation steps:
-
Convert the first 6 hexadecimal digits of
timestampto a decimal number to get 1754393106s, and convert the last 4 hexadecimal digits to a decimal number to get 100us -
Using
1970-01-01 00:00:00as the base, 1754393106 seconds later is2025-08-04 11:25:06. Assuming the time zone is UTC+8 Beijing time, the actual readable time is2025-08-04 19:25:06, zero 100us. That is, the time of the even VCSEL partition pixel points in thisMSOPpacket is:2025-08-04 19:25:06, zero 100us -
Convert the
vcsel_intervalfield to a signed decimal number to get -67us -
According to the formula \(100\mu\text{s} + 256\mu\text{s} - 67\mu\text{s}\), the result is 289us. That is, the time of the odd VCSEL partition pixel points in this
MSOPpacket is:2025-08-04 19:25:05, zero 811us
The EM4-F obtains reflectivity information by parsing the intensity field of individual pixels in MSOP packets. The output is formatted as reflectivity partitions ranging from 0 to 255, where lower partition numbers indicate lower reflectivity and vice versa.
For example
The intensity field of the second pixel in an MSOP packet is read as 6E;
Hexadecimal 6E converts to decimal 110, indicating that the reflectivity of the second pixel falls into bin 110.
- Yaw angle (\(\theta\)):
EM4-Tobtains the yaw angle (\(\theta\)) by parsing theyaw_anglefield (two's complement form) in theMSOPpacket and compensating/correcting it according to theHalfVcselYawOffsetfield (two's complement form) in theDIFOP2packet, with the unit 0.01° - Radial distance (\(r\)): Parse the
radiusfield of a certain pixel point in theMSOPpacket to obtain the radial distance \(r\), with the unit mm - Pixel spacing (PixelPitch): Parse the
PixelPitchfield (two's complement form) corresponding to the pixel point in theDIFOP2packet, and compensate/correct it according to theSurfacePitchOffsetfield (two's complement form) to obtain the Pitch angle (\(\varphi\)) information, with the unit 0.01°
The \(yaw(j, i)\) and \(pitch(j, i)\) of the \(i\)-th pixel point in the \(n\)-th column on the \(j\)-th (\(j = \text{A, B, C, D}\)) rotating mirror surface are equal to:
The Cartesian coordinates x, y, and z are obtained, with the coordinate transformation shown in Figure 12.

Figure 12 MSOP Point Cloud Coordinate
The computational formula is given below::
Known conditions:
- The value read from the
yaw_anglefield in a certainMSOPpacket of mirror A is:15 50 - The value read from the
radiusfield of the 22nd pixel is:00 03 9E A5 - The value read from Byte2-3 corresponding to the second partition of
HalfVcselYawOffsetin theDIFOP2packet is:FF FD - The value read from Byte42-43 corresponding to the 22nd pixel in the
PixelPitchfield is:F9 F6 - The value read from Byte0-1 corresponding to mirror A in the
SurfacePitchOffsetfield is:00 0F
Parameter conversion (hexadecimal two's complement to decimal):
- The hexadecimal two's complement
15 50is converted to the decimal number \(+5456\), andFF FDis converted to the decimal number \(-3\) (unit: 0.01°), so the \(yaw \text{ angle} = 54.53^\circ\) - The hexadecimal number
00 03 9E A5is converted to the decimal number \(263,197\text{ mm}\), so the radial distance \(r = 263.197\text{ m}\) - The hexadecimal number
F9 F6is converted to the decimal number \(-1546\), and00 0Fis converted to the decimal number \(+15\), so the \(pitch \text{ angle} = -15.31^\circ\)
The calculated \((x, y, z)\) coordinate information of this pixel is as follows:
5 Fault diagnosis
This chapter lists some common problems encountered during the use of the product and corresponding troubleshooting methods. Please refer to Table 15 for details.
Table 15 Common Troubleshooting Methods
| Fault phenomenon | resolvent |
|---|---|
| The red/green indicator light on the interface box is not on/flashing | Check if the connection wire between the interface box and the power end is loose; Check if the wiring harness is damaged. |
| The product keeps restarting during startup |
Check if the input power connection and polarity are normal; Check whether the voltage and current of the input power supply meet the requirements (under the condition of 12V voltage input, the input current is ≥ 2A); |
| Wireshark can receive data, but RSView does not display point clouds |
Turn off the computer firewall and run RSView to pass through the firewall; Confirm that the IP configuration of the computer matches the destination address set for the product; Confirm that the Sensor Network Configuration setting in RSView is correct; Confirm that the RSView installation directory or configuration file storage directory does not contain any Chinese characters; Confirm that the received packet in Wireshark is of MSOP type. |
| Frequent data loss of products |
Confirm whether there are a large number of other network packets or network conflicts in the network; Confirm whether there are other network products in the network that are sending large amounts of data in broadcast mode, causing sensor data blockage; Confirm whether the performance and interface performance of the computer meet the requirements; Remove all other network products and connect directly to the computer to confirm if there is any packet loss. |
| Unable to synchronize gPTP time |
Confirm whether the radar firmware matches the required synchronization mode; In gPTP time synchronization mode: Confirm whether the gPTP Master synchronization protocol complies with the current gPTP protocol; Confirm if gPTP Master is working properly. |
| The product has no data output after passing through the router | Turn off the DHCP function of the router or set the sensor's IP address to the correct IP address inside the router. |
6 Product maintenance
6.1 Transportation and Logistics
Improper transportation can cause product damage
- The product should be packaged with shockproof and moisture-proof materials to avoid damage during transportation. It is recommended to use the original packaging;
- Handle with care during transportation to avoid impact or dropping;
- When receiving the goods, carefully check the delivery list for any damages (including the product and packaging);
- If there is any transportation damage, refuse to accept the delivery and contact RoboSense promptly.
6.2 Storage
Improper storage may cause product damage
- Store the product in an indoor environment with normal temperature and dry conditions;
- Handle the product gently to avoid impact or dropping;
- The product should be stored in a safe environment to avoid corrosion, mechanical impact, or exposure to environments exceeding the protection level;
- Regularly inspect the condition of all components and packaging, and it is recommended to check every three months.
6.3 Product Cleaning
To ensure accurate perception of the surrounding environment, keep the RS-LiDAR's circular protective cover clean.
6.3.1 Precautions
Note
- Before cleaning the RS-LiDAR, carefully read and understand the content of this section. Improper cleaning may damage the product.
- When using the LiDAR in harsh environmental conditions, clean the surface regularly to keep the LiDAR clean. Otherwise, it may affect the normal operation of the LiDAR.
6.3.2 Required Materials
- Clean and dust-free cloth;
- Neutral solution at moderate temperature (such as soapy water, distilled water, 99% concentration of ethanol, etc.).
6.3.3 Cleaning Method
-
If the LiDAR surface is only covered with some dust:
- Use a clean and dust-free cloth, dip it in a small amount of neutral solution;
- Gently wipe the LiDAR surface;
- Dry it with a clean and dry dust-free cloth.
-
If the LiDAR surface is covered with mud or other solid foreign objects:
- First, spray clean water on the dirty part of the LiDAR surface to remove the mud or foreign objects (Note: Do not directly wipe off the mud with a dust-free cloth, as it may scratch the surface, especially the protective cover);
- Then spray warm soapy water on the dirty part. The lubricating effect of the soapy water helps to remove the foreign objects. Gently wipe the LiDAR surface with a fiber cloth, but be careful not to scratch the surface;
- Finally, rinse off the residual soap on the LiDAR surface with clean water (if there is still residue, clean it again with 99% ethanol) and dry it with a clean and dry dust-free cloth.
7 After sales
If the solutions provided in Chapter 6 of the troubleshooting guide do not solve the problem, please promptly contact RoboSense.
Official Website: https://www.robosense.cn/en/contact
Email: support@robosense.cn
Phone: +86-0755-86325830 / +86-15338772453
Info
- Please wait for a confirmation response from RoboSense after-sales service before sending the product back;
- When sending the product back, please use the original packaging or an equivalent cushioned and moisture-resistant packaging.
Appendix A Definition of TE connector pins

| Definition of TE connector pins | ||
|---|---|---|
| Pin Number | Definition | Connector model |
| 1 | POWER | TE 2387351-1 |
| 2 | Wakeup | |
| 3 | / | |
| 4 | GND | |
| 5 | / | |
| 6 | / | |
| D1 | 1000BASE-T1 N | |
| D2 | 1000BASE-T1 P | |
