CL-IT-005: Intelligent Connected Passenger Car Advanced Development Platform

I. Platform Overview

This platform is an innovative teaching platform for the development of intelligent connected large passenger cars, designed to improve educational quality and cultivate professional talent in the field of future autonomous driving technology. By performing professional by-wire modifications on mass-produced passenger cars and integrating advanced autonomous driving systems, this platform is capable of achieving Robot Taxi-level automation. It provides students with a hands-on teaching environment to deeply understand and master key by-wire technologies and autonomous driving techniques.

Technical Features

· By-Wire Chassis Technology: The platform uses industry-leading by-wire VCU (Vehicle Control Unit) technology to modify mass-produced passenger cars for excellent by-wire performance. This includes leading industry standards in control precision, error management, response time, and feedback accuracy.

· Functional Safety Design: The design fully considers functional safety, achieving manual takeover capabilities for the accelerator pedal, brake pedal, and steering wheel, ensuring a smooth transition from autonomous driving mode to conventional driving mode. Moreover, the system can automatically switch to conventional driving mode in case of power failure and integrates an emergency power-off device, adding multiple safety guarantees such as redundant control verification and software restrictions on high-speed steering, to ensure driving and vehicle safety.

· Autonomous Driving Software and Sensor Integration: The platform utilizes a full set of self-developed autonomous driving software, covering advanced perception algorithms and obstacle behavior prediction functions. Sensor applications are extensive, including cameras, lidar, millimeter-wave radars, etc., combined with accurate positioning algorithms, such as a laser positioning main strategy supplemented by RTK positioning, to ensure accurate vehicle operation even in complex environments. The platform also supports customized planning and control algorithms for mixed traffic conditions, realizing various autonomous driving functions such as active tracking, obstacle recognition, and active braking.

I. Detailed Explanation of By-Wire Chassis Technology

Overall

Vehicle Control Mode The platform's overall vehicle control mode utilizes advanced by-wire technology, achieving flexible switching of autonomous driving modes. Through precise protocol control, the vehicle's autonomous driving mode can be conveniently activated or deactivated, while supporting unified or independent control of braking, throttle, and steering, providing an enable switch for vehicle speed under automatic control mode, ensuring safety and stability during high-speed driving.

1. By-Wire Throttle

· Functionality and Performance: By-wire throttle technology realizes longitudinal driving control through precise by-wire management, with a CAN control interface supporting precise control of throttle pedal opening (percentage) and torque (Nm).

· Response Characteristics: The by-wire throttle is designed for rapid response, with a delay time of under 300ms from command issuance to acceleration onset, and response time under 800ms to reach maximum acceleration, not less than 0.3g, meeting the demands of high-performance autonomous driving.

· Operational Range and Enablement: Operational speed range from 0 to the maximum speed allowed by the original vehicle, with throttle capable of individual enablement and takeover, allowing manual takeover via the throttle pedal, enhancing operational flexibility and safety.

· Status Feedback: The system provides feedback on by-wire throttle status, actual throttle pedal position, command values, and requested wheel-end torque for precise monitoring and adjustment.

2. By-Wire Brake

l Functionality and Performance: By-wire brake technology achieves high-precision control of longitudinal driving braking function, also with a CAN control interface supporting precise adjustment of brake pedal opening (percentage) and deceleration (m/s^2).

l Performance Requirements: The braking system can provide deceleration of not less than 0.5g, ensuring rapid and safe vehicle stopping in emergency situations.

l Response Characteristics: Delay time controlled under 300ms from command issuance to the onset of deceleration, and response time under 800ms to reach maximum deceleration, meeting the demands for rapid response in emergency braking.

l Operational Range and Enablement: Operational speed range covers from 0 to the maximum speed allowed by the original vehicle, with by-wire braking also supporting individual enablement and takeover, providing manual takeover via the brake pedal, improving driving safety.

l Status Feedback: The system feeds back by-wire brake status, actual brake pedal position, command values, and deceleration command values, providing data support for precise control and fault diagnosis of the braking system.

3. By-Wire Steering

l Functionality and Performance: The platform's by-wire steering technology utilizes high-precision control strategies, with a dedicated by-wire CAN control interface for accurate control of steering wheel angle (degrees) and torque (Nm).

l Performance Requirements: Supports accurate steering control within the speed range of 0 to the maximum speed allowed by the original vehicle. Steering wheel angle range from -470 degrees to 470 degrees, with a maximum steering rate of not less than 1020 degrees per second, meeting the requirements for rapid steering.

l Response Characteristics: Delay time is controlled within 300ms from the issuance of the command to the change in steering angle, ensuring an immediate response to the driver's operation or the autonomous driving system's control commands.

l m. Enablement and Takeover: By-wire steering can be individually enabled and taken over, supporting advanced functions that allow manual takeover via the steering wheel.

l Status Feedback: The system provides feedback on by-wire steering status, actual steering wheel angle, and command values, facilitating the monitoring and optimization of steering control performance.

4. By-Wire Gear Selection

l Function Description: The by-wire gear selection function allows for automatic switching between P, R, N, D gear positions while the vehicle is stationary, using a CAN interface. The functionality can be flexibly enabled or disabled through protocols.

l Safety Requirements: In the automatic control mode, the gear enable switch ensures that the D position can enter or be prohibited from entering by-wire mode as needed, increasing driving safety

l Response Time: The time from command issuance to gear shift completion is controlled within 2 seconds, ensuring a quick and smooth transition

l Status Feedback: Provides feedback on by-wire gear status, actual gear lever position, and command values, supporting precise control and fault detection.

5. By-Wire Lighting and Electronic Parking Brake

l Control Interface: Includes switch control for left and right turn signals, high and low beam headlights, emergency flashers, horn, windshield wipers, as well as electronic parking brake control interface, operated via respective CAN control interfaces.

l Function Implementation: The by-wire implementation of these functions increases flexibility and safety during autonomous driving, supporting automatic operation in complex environments.

6. ▲ Status Feedback

l Comprehensive Information Feedback: The platform provides key information feedback via CAN bus signals, including wheel steering, wheel speed, takeover reasons, vehicle speed, longitudinal acceleration, lateral acceleration, yaw rate, door status, seat belt status, key status, Ultrasonic Sensor ranging distance, external temperature, and total odometer, providing comprehensive data support for system monitoring and fault diagnosis.

7. By-Wire Mode

l Manual and Automatic Modes: The by-wire system's working modes include manual and automatic modes. In manual mode, the vehicle is completely controlled by the driver. In automatic mode, one or more by-wire subsystems of the vehicle are controlled by the intelligent driving computer. The entire vehicle by-wire enable flag bit controls the switch between manual and automatic modes.

l Default Mode and Fault Switching: The system defaults to manual driving mode upon power-up and automatically switches to conventional driving mode in the event of a severe fault, ensuring driving safety.

8. By-Wire Takeover

l Takeover Settings: The system can be set to enable a smooth takeover from by-wire mode to manual mode via any one or several of the throttle, brake, or steering, ensuring quick restoration of driver control in emergency or specific situations, enhancing safety and flexibility of driving.

II. Autonomous Driving System Configuration

1. Computing Unit

· Core Configuration: The platform's computing unit is equipped with a high-performance 6-core 12-thread processor, with a clock speed of 2.9GHz, complemented by 12M of L3 cache, and an independent graphics processing unit with 3584 CUDA processors, a memory frequency of 15Gbps, and a 12G DDR6 memory capacity.

· Memory and Storage: The system memory is configured with 16GB LPDDR4x at a frequency of 2666MHz, paired with a 500GB solid-state drive, ensuring fast data processing and storage capabilities.

· Networking and Interfaces: Provides gigabit Ethernet and WiFi connection options, as well as USB 3.0 interface, supporting fast data transfer and peripheral connections.

9. Mid-range Front-facing Cameras x2

· Technical Specifications: Two mid-range front-facing cameras, using the OV2775/RGB/RS chip, sensor size 1/2.9 inches, pixel size 2.8um, with built-in IPS chip, output resolution of 1920x1536, supporting hardware synchronization.

· Viewing Angle and Lens: Each camera has a horizontal field of view (HFOV) of 60 degrees, vertical field of view (VFOV) of 46 degrees, focal length of 5.8mm, waterproof rating of IP67, ensuring reliable environmental adaptability.

10. Long-range Front-facing Camera

· Technical Specifications: The long-range front-facing camera also uses the OV2775/RGB/RS chip, sensor size 1/2.9 inches, pixel size 2.8um, with built-in IPS chip,output resolution of 1920x1536, providing hardware synchronization.

· Viewing Angle and Lens: This camera has a horizontal field of view (HFOV) of 30 degrees, a vertical field of view (VFOV) of 24 degrees, and a focal length of 11mm. It also has a waterproof rating of IP67, designed for long-range vision.

11. Surround-view Cameras

· Technical Specifications: The surround-view cameras use the ISX021/RGB/RS chip, sensor size 1/2.57 inches, pixel size 3um, with built-in IPS chip, output resolution of 1920x1080, supporting hardware synchronization.

· Viewing Angle and Lens: They provide a horizontal field of view (HFOV) of 118 degrees and a vertical field of view (VFOV) of 92 degrees, with a focal length of 3mm and a waterproof rating of IP67. These cameras provide a 360-degree panoramic view around the vehicle, enhancing environmental perception during autonomous driving.

12. 80-line Lidar

· Performance Parameters: Uses 80-line lidar with a wavelength of 905nm, offering a maximum detection range of 200m and high accuracy of ±2cm. Frame rates can reach up to 20Hz, with a broad operating temperature range from -30°C to +60°C, adaptable to various environmental conditions.

· Application Functions: This lidar provides a 360-degree spatial scanning capability for autonomous vehicles, supporting obstacle detection, classification, and tracking in complex environments, and is a key component for achieving advanced autonomous driving functions.

13. Combined Positioning Unit

· Technical Specifications: The combined positioning unit supports RTK and GNSS single-point modes, using tri-mode, seven-frequency positioning (GPS, BDS, GLONASS), and includes a 6-axis IMU (Inertial Measurement Unit), providing high-precision location and posture information for the autonomous driving system.

· Positioning Performance: By combining high-precision satellite navigation with ground-based augmentation systems, it ensures stable and accurate positioning services in various environments, which is crucial for route planning and navigation.

14. Millimeter-wave Radar

· Operating Frequency and Detection Range: Millimeter-wave radar operates within the frequency range of 76GHz to 77GHz, with a detection range from 0.2m to 250m, covering a wide range of target detection distances.

· Performance Features: It has a long-range resolution and accuracy of ±1.79m and a short-range of ±0.39m; with a measuring accuracy of ±0.40m at long distances and ±0.10m at short distances. The detection angle range reaches 120°, and the speed range is from -400 km/h to +200 km/h, supporting the measurement of various targets' speed and distance.

· Data Output and Environmental Adaptability: Provides CAN/CANFD data output, including tracking target ID, distance, speed, RCS, etc. The working temperature range is from -40℃ to 85℃, with an operating voltage of 9-16V, and a protection level not less than IP67, ensuring reliability and stability in all weather and road conditions.

15. Ultrasonic Sensor

· Measuring Capacity and Accuracy: The Ultrasonic Sensor has a measuring range from 130mm to 5000mm, with a blind zone of 13cm. The beam angle is adjustable from 10° to 60°, with an accuracy of 5mm at close range and 0.5% of the measured distance at long range.

· Operating Conditions and Function: The processing board and probe have an operating temperature range from -40℃ to 85℃, operating power of +12V to +24V, and operating current less than 200mA. Ultrasonic Sensor is mainly used in autonomous driving systems for close-range obstacle detection and avoidance, particularly important in low-speed driving and parking scenarios.

III. Platform Functions  

This autonomous driving teaching platform integrates a series of advanced functions, aiming to provide a comprehensive autonomous driving experience and teaching support:

1. Complete Autonomous Driving System: The platform is equipped with a complete autonomous driving system that can perform L4-level autonomous driving functions under various road conditions, including but not limited to active tracking, obstacle recognition, active braking, precise docking at stops, and dynamic local route planning.

2. High-precision Map Generation and Use: The platform supports the generation of high-precision map information sources, capable of recording point cloud data packages and using map-making software to create detailed high-precision maps, which are crucial for route planning and positioning accuracy.

3. Independent Sensor Applications: To deepen the learning experience, the platform provides specialized training software for individual sensors, allowing learners to understand and practice the working principles and application scenarios of each sensor.

4. Integration of Various Positioning Technologies: Combining various positioning technologies, the platform can achieve precise tracking or navigation based on high-precision maps in both indoor and outdoor environments, enhancing the adaptability and flexibility of the autonomous driving system.

IV. Supportive Software 

To fully utilize the platform's potential for teaching and research, the following supportive software has been carefully designed to provide comprehensive learning and development support:

1. Visual Testing Software:

· Includes modules for vehicle and pedestrian recognition, lane line detection, and traffic light recognition.

· Supports the fast installation, calibration, and debugging of cameras, as well as data collection and processing.

· Provides a complete development toolkit, including Python3, TensorFlow, CUDA, etc.

· Contains machine learning models, training samples, and real-time processing demos, complemented by training datasets.

Radar Testing Software:

· Designed for testing millimeter-wave and Ultrasonic Sensors, including distance detection and target detection.

· Capable of receiving and analyzing radar data streams, suitable for target analysis under different operating conditions.

· Offers a function for reading fault information.

Lidar Testing Software:

· Supports interface testing and lidar configuration (including network settings, time synchronization, motor parameter adjustments, etc.).

· Real-time reception of lidar data streams, visualizing point cloud information.

Combined Navigation Testing Software:

· Includes interface testing and calibration of the combined navigation system (initial alignment, navigation mode, coordinate axis configuration, data output, etc.).

· Supports reception and analysis of combined navigation data, providing a function for reading fault information.

These components and software suites create a robust framework for students and researchers to engage deeply with autonomous driving technologies, facilitating an understanding of both the operational aspects and the challenges of developing autonomous driving solutions.