ADCU - Autonomous Driving Control Unit
Introduction
Ecotron ADCU is a cutting-edge platform with powerful computational capability tailored for autonomous driving system. The ADCU receives data from multiple sensors, e.g. cameras, radars, LiDAR, and cloud data (V2X), and acquires vehicle dynamic data from VCU so that it can support decision-making and route planning algorithms. The outputs of the ADCU will be used for the driving status feedback, vehicle control, and various autonomous driving features. Ecotron ADCU can be customized according on customer vehicle parameters and components input specifications.
Functions
EAORA04 – Based on Nvidia Jetson Orin
Designed to Support L4 Autonomous Driving
EAORA04-D is an intelligent computing platform developed by Ecotron, for autonomous driving systems using NVIDIA Jetson Orin chip and Infineon TC297. Using the supporting basic software and development tools, developers can build an L4 low-speed autonomous driving system in a safe, convenient, and efficient manner.
NVIDIA Jetson Orin is designed for embedded intelligent systems including autonomous driving systems which can be used to implement autonomous driving functions such as sensor fusion, environment perception, and path planning, etc. Infineon TC297 is based on TriCoreTM architecture with a 300MHz operating frequency and an ECC (Error Correction Code) protected RAM with 728KB + 8MB capacity. It is designed based on the ISO26262 standard and supports ASIL-D safety level requirements. Developers can develop and deploy vehicle control and functional safety related strategies based on the MCU.
EAXVA05 - Based on Dual NVIDIA Jetson Xavier
Designed to Support L4 Autonomous Driving
Ecotron’s new generation ADCU, EAXVA05, is an intelligent computing platform designed specifically for autonomous driving systems. EAXVA05 is equipped with 2 NVIDIA Xavier chips and 1 Infineon TC297 chip. With basic software and development tools, developers can safely, conveniently, and efficiently build low-speed L4-level autonomous driving systems in confined areas.
NVIDIA designs Xavier specifically for embedded intelligent systems. Xavier supports autonomous driving functions such as sensor fusion, environment perception, and path planning. Infineon TC297 has a TriCore™ architecture and has an operating frequency of 300MHz. Moreover, it has 728KB + 8MB capacity and ECC (error correction coding) RAM protection. Ecotron’s ADCU is designed based on the ISO26262 standard and supports ASIL-D safety level requirements. Engineers can develop vehicle control and functional safety-related strategies based on this MCU.
EAXVA04 - Based on NVIDIA Jetson Xavier
Designed to Support L3/L4 Autonomous Driving
Powered by NVIDIA Xavier, Ecotron ADCU EAXVA04 targets for L3/L4 Autonomous Driving. EAXVA04 is not only able to perceive and interpret the vehicle surrounding environment via different kinds of sophisticated sensors with decent processing speed and high accuracy, but also, based on the acquired data, implement advanced AI algorithms and vehicle control strategies to implement autonomous driving.
Essentially, ADCU platform EAXVA04 provides a solution making it possible to integrate ADAS/ autonomous system design/ implementation and vehicle control strategy design/ implementation within a single unit, greatly enhancing the overall system integrity. In addition, multiple types of communication ports meet the requirements of most autonomous driving sensors available on market nowadays - LVDS channels of the ADCU can be used for vision-based applications; CAN bus and Ethernet ports provide capabilities for integrating radar, LiDAR and all other CAN or Ethernet-based sensors.
Local Development Tool Kit
Ecotron ADCU has a set of development tools installed on the local target, including gcc, make, CMake, catkin, Bazel and gdb debugger. Application developers can develop user space applications directly on the target controller.
EcoCoder-AI, Automatic Code Generation Tool
EcoCoder-AI is a powerful automatic code generation library added on MATLAB/Simulink that links directly to the target controller. EcoCoder-AI integrates automatic code generation, compilation and one-click generation of executable files. It can directly convert the Simulink models into an executable program for the target controller and download to the target controller. It enables the use of Simulink Library, and greatly facilitates controls engineers to graphically develop applications based on autopilot middleware.
Also, EcoCoder-AI enables the use of Simulink external mode to do on-the-fly calibration.
Cross-Compilation Tool
Simulink External Mode
EAORA04 (Orin + TC297)
- Nvidia Jetson Orin & TC297
- 200 TOPS
- 8 CAN buses (Support Radar input)
- 8 HD Cameras
- Multiple I/O interfaces
EAXVA05 (Xavier + Xavier + TC297)
- NVIDIA Jetson Xavier * 2 & TC297TP
- 64 TOPS
- 10 CAN buses (Support Radar input)
- 12 HD Cameras
- Multiple I/O interfaces
EAXVA04 (Xavier+ TC297T)
| Interface Type | EAORA04 Orin + TC297 |
EAXVA05 Xavier*2 + TC297 |
EAXVA04 Xavier + TC297T |
| Video input | 8 channels, FPDLink III | 12 channels, FPDLink III | 8 channels, FPDLink III |
| Video output | 1 channel HDMI | 2 channels HDMI | 2 channels, 1 channel HDMI, 1 channel FPDLink III |
| USB | 1 channel, USB Host support USB2.0, US3.0 and USB3.1 |
2 channels TYPE-A | 4 channels, 3 channels USB Host support USB2.0, US3.0 and USB3.1, 1 channel USB Device for debug |
| RS232 | 4 channels, 1 channel for debug |
8 channels, 1 channel for debug |
4 channels, 1 channel for debug |
| Automotive Ethernet | 3 channels, 100Base-T1/1000Base-T1 |
5 channels, 100Base-T1/1000Base-T1 |
10 channels, 1000Base-T1, 100Base-T1 adaptive |
| Standard Ethernet | 2 channels, 100BASE-T/1000BASE-T |
4 channels, 100BASE-T/1000BASE-T |
5 channels |
| CAN | 2+6 channels, 2 channels connected to SOC, 6 channels connected to MCU, CAN FD compatible, 2 of them with specific frame wake-up |
4+6 channels, 4 channels connected to SOC, 6 channels connected to MCU, CAN FD compatible, 2 of them with specific frame wake-up |
2+6 channels, 2 channels to SOC, 6 channels connected to MCU, CAN FD compatible, 2 of them with specific frame wake-up |
| LIN | 4 channels | 4 channels | 2 channels |
| Digital input | 6 channels, 4 channels active high, 2 channels active low |
6 channels, 4 channels active high, 2 channels active low |
20 channels, 10 channels active high, 6 channels active low, 4 frequency inputs |
| Analog input | 6 channels, 2 channels 5V voltage type, 2 channels 36V voltage type, 2 channels resistance type |
6 channels, 2 channels 5V voltage type, 2 channels 36V voltage type, 2 channels resistance type |
13 channels, 6 channels 5V voltage type, 4 channels resistance type, 3 channels 36V voltage type |
| Digital low-side output | 8 channels@250mA | 8 channels@250mA | 18 channels, 4 of them can be configured as PWM outputs |
| Digital high-side output | 4 channels@1A | 4 channels@1A | 10 channels, 2 of them can be configured as PWM outputs |
| 5V sensor power supply | 2 channels maximum current 100mA | 2 channels maximum current 100mA | 5 channels, 2 channels maximum current 100mA, 3 channels maximum current 50mA |
