Reconnaissance payload on interchangeable wheeled/tracked chassis with 3D LiDAR mapping, autonomous navigation, and real-time sensor data relay

Built for field training, emergency rescue, remote-area patrol and hazardous-zone reconnaissance, the system sends an unmanned platform into the target area in place of personnel. It follows a "recon payload + interchangeable chassis + robot software" architecture: the payload integrates 3D LiDAR, depth camera, observation PTZ, front camera, IMU, differential positioning and an onboard computer into a self-contained sensing and computing unit; wheeled and tracked chassis options match the mission terrain, and the payload mounts across both. On site, the robot is ready after power-on: teleoperate to scan and map, then it autonomously plans paths, avoids obstacles and reaches designated recon positions, relaying sensor data to the operator terminal in real time.
| System | Recon payload + interchangeable wheeled/tracked chassis |
| Platform (typical) | 930×699×348mm, 62kg, 50kg payload, 6km/h, 135mm obstacle, ≤30° slope, IP22 |
| 3D LiDAR | 32-line, 0.3-20m range, 90° vertical / 360° horizontal FOV, 10/20Hz, Class 1 eye-safe |
| Depth Camera | 1MP ToF depth + 12MP RGB, built-in 3-axis accel/gyro, 7-mic array |
| Observation PTZ | 0.05° precision, ±60° pan/tilt, 30°/s max, 1kg payload |
| Positioning | Diff-GPS + IMU, 0.5° heading (RMS), 0.001° pitch/roll, ±2000°/s gyro |
| Front Camera | 1280×720@30fps with HDR |
| Onboard Computer | 8-core Arm Cortex-A78AE, NVIDIA Ampere GPU (1792 CUDA / 56 Tensor cores), 32GB RAM / 512GB SSD |
| Software | ROS-based: LiDAR mapping, autonomous localization & navigation; standard point cloud / image / odometry / velocity interfaces |
| Data Relay | Real-time relay of RGB / depth imagery, point clouds and robot pose to the operator terminal |
| Power | 30Ah Li-ion battery, 2.5-3h charging |
| Environment | 0-45°C, <80% RH (non-condensing) |
Turn the chassis Q4 switch and release the side e-stop, then press the payload power button. The payload carries all sensors and computing needed for mapping and navigation; the chassis executes motion tasks.
After launching the ROS bringup, teleoperate the robot to scan the scene; gmapping builds the map and map_server saves it. The robot then navigates autonomously to goal points, with MoveIt arm planning available as an extension.
Standard ROS interfaces cover chassis and base sensors, localization/navigation, imagery and point clouds (point cloud / image / odometry / velocity). The back panel exposes 2× USB 3.0, Ethernet and HDMI for secondary development.
Operate at 0-45°C, <80% RH non-condensing. Never hot-plug components while powered. Recharge promptly on low battery and every two months in storage; store batteries at -5-35°C in a clean, dry, ventilated place away from heat sources.
Excerpted from the Robot Payload User Manual V1.1. Refer to the delivered manual for the full safety warnings, electrical guidelines and development guide.