FXB-Z04003 Apollo Self-driving Development Platform

About Feature Technical Parameter Basic Configuration

Apollo self-driving training platform adopts Apollo self-driving technology to develop. The training platform integrates wire-controlled chassis, LIDAR, millimeter wave radar, camera, combined inertial navigation and other environmental perception sensors, high-performance computing unit. It can realize RTK tracing, perception and obstacle avoidance, and other self-driving functions. The system covers wire-controlled, perception, prediction, planning, control and other knowledge points and applications related to self-driving. And it provides developer tools, which is suitable for intelligent connected vehicles (ICV) technology training needs.

1. Industrial integrated structure design, easy to disassemble and operate, convenient for equipment debugging and troubleshooting.
2、Pure wire-controlled design, integrated control, specially built for self-driving, providing the whole vehicle CAN interface protocol and framework system such as Apollo for self-driving, supporting users' secondary development and all-round capability cultivation.
3. The vehicle body is equipped with 1 piece of binocular HD camera, 8 piece of vehicle-level ultrasonic radar, 1 piece of 77GHz millimeter wave radar. Through the self-driving system to achieve all-round obstacles avoidance to provide safer driving.
4. Provide the operation manual of self-driving kit, the manual of self-driving kit software.
5. The following system calibration functions are available: vehicle speed correction, left turn light correction, right turn light correction, camera height data setting, camera center position setting in front windshield, camera to wheel length setting, vehicle body width setting, camera to vehicle head length setting, camera angle adjustment setting, lane center line calibration, vehicle hood line setting, activation speed calibration, lane line sensitivity calibration, alarm tone type setting, forward collision warning sensitivity calibration, etc.
6. With emergency stop switch, front and rear crash bar, command verification, heartbeat protection, steering system fault handling, drive system fault handling, emergency power-down parking protection, battery fault monitoring, online detection of the whole vehicle CAN node, whole vehicle fault level division processing, vehicle fault alarm, vehicle emergency deceleration prompt, vehicle loss of control protection, remote control off-line processing, charging safety monitoring and protection, remote control distance ≥ 100m.
7. With binocular camera calibration function, the following calibration functions can be realized: calibration of value from left camera to left front windshield edge, calibration of value from left camera to right front windshield edge, calibration of value between the outside edge of two front wheels, calibration of value from left camera to ground, calibration of value from left camera to front bumper, setting of value from car front to ground, target learning, optical parallax learning.
8. With GPS combined navigation calibration function, calibration of combined navigation own parameters, including calibration of value from positioning antenna to the rear wheel center rod arm, calibration of value from inertial navigation to GNSS positioning antenna rod arm, calibration of GPS inertial navigation system motion initialization configuration.
9. With LIDAR configuration and test function, the LIDAR can be configured and tested by using the upper computer software.
10. With millimeter wave radar configuration and test function, the forward detection angle and detection target type of millimeter wave radar can be set by using the upper computer software.
11. With industrial computer configuration function, the IP address between industrial computer and sensor can be set, the CAN network communication data, serial port, chassis, LIDAR point cloud, imu signal, GPS signal and other data can be viewed.
12. With chassis adaptation function, it can complete the adaptation of new chassis in CAN BUS module.
13. With self-driving function verification, including map acquisition recording function, tracing demonstration verification, obstacles parking demonstration verification, obstacles bypassing demonstration verification.
14. Use mainstream pure electric vehicle power battery, lithium iron phosphate power battery, single battery 3.2V 20Ah, a total of 15 cells in series, total voltage 48V. Distributed battery management system, 2 units of collection modules, each collection module is responsible for multiple single battery information collection. 1 unit of master control module, master control module communicates with 2 units of collection modules through CAN network. The power battery output high-voltage electricity to different electric drive systems.
15. Equipped with mainstream pure electric vehicle single-gear transmission, drive shaft, electric brake and other devices, which can realize the actual working conditions such as downhill driving, flat road driving, uphill driving, half-hill starting, turning left or right, etc.
16. Built-in DC/DC converter, for converting power into 48V, 12V, 24V.

1. Chassis and bracket
1.1 Overall dimension (mm): not less than 1315 × 765 × 1420(length × width x height),including front and rear collision protection device and sensor bracket
1.2 Maximum speed: not less than 8km/h
1.3 Minimum distance from the ground: not less than 100mm
1.4 Minimum turning radius: less than 2000mm
1.5 Overall bridge type suspension, wheelbase 660mm, wheel track 645mm, tire diameter 420mm.

2.Power battery
2.1 Battery power: not less than 20AH
2.2 Charging time: less than 5h
2.3 Battery voltage: 48V
2.4 endurance: not less than 3h
2.5 Support for power battery exchange

3.Equipped with computing platform
3.1Graphics performance:
GPU: 512 Volta architecture CUDA
Processor performance CPU: 8-core ARM64 architecture 2.26GHz (4x2MB L2 + 4MB L3)
Hard disk: 500GB SATA SSD
3.2 Equipped with auxiliary tool: high-brightness monitor not less than 15.6 inches, monitor bracket, interface expansion panel

4. Equipped with combined navigation sensor
Attitude accuracy: 0.1 degree ( baseline length not less than 2m)
Positioning accuracy:
Single point L1/L2: 1.2m
DGPS: 0.4m
RTK: 1cm+1ppm
Data update rate: 100Hz
External interfaces:
3× RS232 1 × RS422 1 × CAN 1 × Micro USB interface
2 × GNSS antenna interface 1 × 4G antenna interface
Operating temperature: -40° C ~ +70° C
Protection level: IP67
Input voltage: 9～32V DC
Power consumption: less than 5W (typical value)

5. Equipped with LIDAR
5.1 Number of channels: not less than 16
5.2 Perceived data points per second: not less than 320,000
5.3 Perception range: not less than 150m
5.4 Perception accuracy: less than ±3cm
5.5 Vertical measurement angle range: 30° (± 15°)
5.7 Vertical direction angular resolution: less than 2°
5.8 Horizontal direction angle resolution: 0.09°-0.36°

6. Equipped with millimeter wave radar
6.1 Operating bandwidth: 76GHz-77GHz
6.2 Distance detection accuracy: less than ±0.1m near, ±0.4m far
6.3 Speed detection accuracy: less than ±0.1km/h
6.4 Azimuth angle detection accuracy: less than ±0.3°@0°/±1°@±45°/ ±5°@±60° near, ±0.1° far

7.Equipped with binocular camera
Processing unit: FPGA, dual-core ARM processor, 1GB memory, 8GB flash storage, resolution 1280x720, communication interface: Gigabit LAN port, RS485, CAN.

8. Equipped with wire-controlled chassis, the VCU of wire control chassis:
Main frequency 168MHz, flash 512kb hardware floating point acceleration, CAN interface, CAN2.0B, kinematic analysis.

9. Equipped with special camera calibration plate with bracket, and millimeter-wave radar corner reflector.

10.Programming language: C++, Python

1 unit of 16-wire LIDAR, 1 unit of 77Ghz millimeter wave radar, 1 set of ultrasonic radar, 1 unit of binocular camera, 1 unit of computing platform, 1 set of wire-controlled chassis, 1 set of sensor fixed mount, 1 set of GPS navigation, 1 unit of CAN analyzer, 1 unit of router.