HesaiLidar_SDK_2.0

👉 Chinese version

1 Check Compatibility

1.1 Lidar Models

Pandar	OT	QT	XT	AT	FT	JT
Pandar40P	OT128	PandarQT	PandarXT	AT128E2X	FT120	JT16
Pandar64	-	QT128C2X	PandarXT-16	AT128P	-	-
Pandar128E3X	-	-	XT32M2X	ATX	-	-

1.2 Operating Systems

• Ubuntu 16/18/20/22.04
• Windows 10

1.3 Compiler Versions

Ubuntu

• Cmake 3.8.0 and above
• G++ 7.5 and above

Windows

• Cmake 3.8.0 and above
• MSVC 2019 and above

1.4 Dependencies

• If using point cloud visualization features, PCL installation is required
• If parsing PCAP files, libpcap installation is required
• If using TLS/mTLS-based Ptcs communication (supported by some lidars), openssl installation is required

2 Getting Started

2.1 Clone

git clone --recurse-submodules https://github.com/HesaiTechnology/HesaiLidar_SDK_2.0.git

On Windows systems, downloading the repository as a ZIP file is not recommended as it may cause compilation errors due to symbolic link issues.

2.2 Compilation

2.2.1 Compilation Instructions for Ubuntu

# 0. Install dependencies
sudo apt update && sudo apt install -y libpcl-dev libpcap-dev libyaml-cpp-dev openssl

# 1. Navigate to source directory
cd HesaiLidar_SDK_2.0

# 2. Create build directory and navigate to build directory
mkdir -p build && cd build

# 3. Configure project with Cmake
#    - Add -DCMAKE_BUILD_TYPE=Release for optimized compilation
cmake -DCMAKE_BUILD_TYPE=Release ..

# 4. Compile SDK
#    - Use -j$(nproc) to utilize all CPU cores
make -j$(nproc)

2.2.2 Compilation Instructions for Windows

Please refer to How to Compile SDK on Windows.

2.2.3 Remove dependency on openssl library (not using PTCS communication)

Please refer to the operations in Compile Macro Control to configure the macro WITH_PTCS_USE to be inactive.

3 Application Guide

3.1 Parse Lidar Data Online

Please refer to How to Parse Lidar Data Online.

3.2 Parse PCAP File Data Offline

Please refer to How to Parse PCAP File Data Offline.

3.3 Point Cloud Data Visualization

Please refer to How to Visualize Point Cloud Data.

3.4 Coordinate Transformation

Please refer to How to Perform Coordinate Transformation.

3.5 Save Point Cloud Data as PCD Files

Please refer to How to Save Point Cloud Data as PCD Files.

3.6 Use GPU Acceleration

Please refer to How to Use GPU Acceleration for Performance Optimization.

3.7 Invoke SDK API Command Interface (PTC Communication)

Please refer to How to Invoke SDK API Command Interface.

3.8 Common Troubleshooting (WARNING)

Please refer to Common Troubleshooting (WARNING).

3.9 Packet Loss Statistics

Please refer to How to Perform Packet Loss Statistics.

3.10 Use Multi-threading to Accelerate Parsing

Please refer to the thread_num configuration in Functional Parameter Reference and configure it to a value >1

Note: The maximum allowed thread count is [CPU maximum cores - 2]. If configured beyond this, it will be modified to this number. Multi-threading will consume more CPU resources, please configure appropriately.

3.11 Parse Multiple Lidar Data Online

Navigate to multi_test.cc

For parsing parameter configuration reference, see How to Parse Lidar Data Online

The basic principle is to use multi-threading to start two SDKs to parse data

3.12 Filter and Parse Specified Lidar Data from PCAP or Real-time Reception Containing Multi-lidar Data

Please refer to the descriptions of device_udp_src_port and device_fault_port in Functional Parameter Reference

Enable point cloud packet filtering by configuring device_udp_src_port (point cloud packet source port number) and device_ip_address (point cloud packet source IP), parsing only point cloud packets from this source IP + source port number.

Enable fault message filtering by configuring device_fault_port (fault message source port number) and device_ip_address (fault message source IP), parsing only fault messages from this source IP + source port number.

3.13 Get Specific Timestamp for Each Point in Pandar Series, OT128, XT Series, QT Series Lidars (Point Cloud Packet Timestamp + Firing Channel Time Correction)

Please use the LidarPointXYZICRTT structure to declare HesaiLidarSdk, where uint64_t timeSecond is the seconds time part and uint32_t timeNanosecond is the nanoseconds time part. For example: HesaiLidarSdk<LidarPointXYZICRTT> sample;

3.14 Set SDK Point Cloud Reception Timeout and PTC Timeout During Initialization

1. Set SDK point cloud reception timeout

   Please refer to recv_point_cloud_timeout in Functional Parameter Reference. This parameter defaults to -1, meaning that during initialization, if no valid point cloud is received, it will block and wait indefinitely. When this parameter is configured to >= 0, the SDK will wait for a period of time before initialization fails and exits.

2. Set PTC timeout

   Please refer to ptc_connect_timeout in Functional Parameter Reference. This parameter defaults to -1, meaning that during initialization, if in DATA_FROM_LIDAR mode, it will block and wait for PTC connection indefinitely. When this parameter is >= 0, the SDK will wait for a period of time before reporting a connection timeout error and continuing initialization.

3.15 Point Cloud Rearrangement Based on Horizontal and Vertical Angles

Please refer to Point Cloud Rearrangement Function

4 Functional Parameter Reference

Please refer to Functional Parameter Reference.