Introduction

Before the commercial application of autonomous vehicles, extensive road testing is required to meet commercial standards. Using road tests to optimize autonomous driving algorithms is time-consuming and costly, and open road testing is still subject to regulatory restrictions. Extreme traffic conditions and scenario replication are difficult, and safety concerns exist during testing. Currently, autonomous driving simulation testing has been widely accepted in the industry,with approximately 90% of autonomous driving algorithm testing completed on simulation platforms, 9% in test fields, and 1% through actual road tests. Autonomous driving simulation testing platforms must possess several core capabilities:realistic scene restoration, efficient use of road data to generate simulation scenarios, and large-scale cloud parallel acceleration to ensure that simulation testing meets the closed-loop algorithms for perception, decision-making, planning, and control in autonomous driving. Currently, various entities, including technology companies, automotive manufacturers, autonomous driving solution providers, simulation software companies, universities, and research institutions, are actively engaged in the construction of virtual simulation platforms.This article provides a detailed introduction to existing autonomous driving simulation software for readers’ reference, and the software ranking is not in any particular order.

01

CarSimCarSim, along with related TruckSim and BikeSim, is a powerful dynamics simulation software developed by Mechanical Simulation Corporation, widely used by manufacturers and suppliers around the world. CarSim is designed for four-wheeled vehicles and light trucks, TruckSim is for multi-axle and dual-tire trucks, and BikeSim is for two-wheeled motorcycles.CarSim is a vehicle dynamics simulation software that primarily simulates from a whole vehicle perspective. It has a considerable number of vehicle mathematical models built-in, all with rich empirical parameters, allowing users to use them quickly without the complex modeling and tuning process.CarSim modelsrun on computers at speeds up to 10 times faster than real-time, simulating the vehicle’s response to driver control, 3D road surfaces, and aerodynamic inputs, with simulation results closely approximating real vehicles, primarily used to predict and simulate the overall vehicle’s handling stability, braking, ride comfort, power performance, and economy. CarSim comes with a standard Matlab/Simulink interface, facilitating co-simulation with Matlab/Simulink for control algorithm development, while generating a large amount of data results during simulation for subsequent analysis or visualization using Matlab or Excel. CarSim also provides an RT version that supports mainstream HIL testing systems such as dSpace and NI systems, enabling convenient joint HIL simulation.

CarSim also supports ADAS-related functions, allowing the construction of parameterized road models, over 200 moving traffic objects, controlling their movement using scripts or through external control in Simulink, while adding up to 99 sensors for detecting both moving and stationary objects. The latest version of CarSim has strengthened its capabilities in ADAS and autonomous driving development, adding more 3D resources such as traffic signs, pedestrians, and the import process of high-precision maps. CarSim also provides an Unreal Engine plugin for co-simulation with the Unreal Engine.

02

CarMakerCarMaker, along with related TruckMaker and MotorcycleMaker, is a dynamics, ADAS, and autonomous driving simulation software launched by the German company IPG. CarMaker is first and foremost an excellent dynamics simulation software, providing accurate vehicle body models (engine, chassis, suspension, drivetrain, steering, etc.). In addition, CarMaker has created a closed-loop simulation system that includes vehicles, drivers, roads, and traffic environments.IPGRoad: can simulate various forms of roads, including multi-lane roads and intersections, and can generate cone-shaped, cylindrical, and other forms of roadblocks through a configured GUI. The geometric shape of the road and surface conditions (unevenness, roughness) can be defined arbitrarily.IPGTraffic: is a traffic environment simulation tool that provides rich traffic object models (vehicles, pedestrians, road signs, traffic lights, road construction, etc.). It can simulate real traffic environments. Test vehicles can recognize traffic objects, triggering actions (e.g., speed limit signs can trigger vehicles to decelerate accordingly).IPGDriver: an advanced, self-learning driver model that can control vehicles under various driving conditions, performing operations such as starting on an incline, parking in a garage, and drifting while turning the steering wheel. It can adapt to the vehicle’s power characteristics (driving type, gearbox type, etc.), road friction coefficients, wind speed, and traffic environment conditions to adjust driving strategies.CarMaker, as a platform software, can integrate with many third-party software such as ADAMS, AVL Cruise, rFpro, etc., leveraging the advantages of each software for co-simulation.Simultaneously, the hardware accompanying CarMaker provides numerous board interfaces, making it easy to conduct HIL testing with ECUs or sensors.

03

PreScanPreScan is an ADAS testing simulation software developed by TassInternational, acquired by Siemens in August 2017. PreScan is a simulation platform consisting of a GUI-based preprocessor for defining scenes and an execution environment for running scenes. Engineers use MATLAB and Simulink as the main interface for creating and testing algorithms. PreScan can be used for applications ranging from Model-in-the-Loop (MIL) control design to real-time testing using Software-in-the-Loop (SIL) and Hardware-in-the-Loop (HIL) systems.PreScancan run in open-loop, closed-loop, offline, and online modes. It is an open software platform with a flexible interface that can connect to third-party vehicle dynamics models (e.g., CarSIM and dSPACE ASM) and third-party HIL simulators/hardware (e.g., ETAS, dSPACE, and Vector).PreScan consists of multiple modules, primarily divided into four steps: building scenes, adding sensors, adding control systems, and running simulations.Scene Building: PreScan provides a powerful graphical editor, allowing users to build rich simulation scenes using road segments, including traffic signs, trees, and buildings from a basic component library, as well as a library of traffic participants, including motor vehicles, bicycles, and pedestrians. Users can modify weather conditions (e.g., rain, snow, fog) and light sources (e.g., sunlight, headlights, streetlights). The new version of PreScan also supports importing high-precision maps in OpenDrive format to create more realistic scenes.Adding Sensors: PreScan supports a wide variety of sensors, including ideal sensors, V2X sensors, LiDAR, millimeter-wave radar, ultrasonic radar, monocular and binocular cameras, and fisheye cameras. Users can add them according to their needs.Adding Control Systems: Control models can be established through MATLAB/Simulink or closed-loop control with third-party dynamics simulation models (e.g., CarSim, VI-Grade, dSpace ASM vehicle dynamics models).Running Experiments: The 3D visualization viewer allows users to analyze experimental results while providing image and animation generation capabilities. Additionally, the interface using ControlDesk and LabView can be used to automatically run batches of experimental scenes and conduct hardware-in-the-loop simulations.

04

PTV VissimVissim is a world-leading microscopic traffic flow simulation software provided by the German company PTV. Vissim can easily construct various complex traffic environments, including highways, large roundabouts, and parking lots, and can simulate the interactions between motor vehicles, trucks, rail traffic, and pedestrians within a simulation scene. Itis an effective tool for planning and evaluating urban and suburban traffic facilities, and can also simulate the impact of localized emergency situations, mass evacuations of pedestrians, and more.Vissim’s simulations can achieve high precision, including microscopic individual following and lane-changing behaviors, as well as group cooperation and conflict. Vissim has built-in multiple analysis methods, capable of obtaining various specific data results under different conditions, and can provide an intuitive understanding through a high-quality 3D visualization engine. Autonomous driving algorithms can also be tested using simulated high-dynamic traffic environments by accessing Vissim.

05

TESS NGThe TESS simulation system is the first generation of road traffic simulation systems developed by Professor Sun Jian at Tongji University in 2006. Since then, over ten years, Professor Sun Jian’s research team has conducted over 100 model innovations and simulation system application practices targeting the operational characteristics of mixed traffic flow in China. The main functions of the TESS NG micro-traffic simulation system include: full traffic scene simulation, multi-mode traffic simulation, intelligent transportation system simulation, visual evaluation, secondary development interfaces, and support for 3D scene display, etc. At the same time, TESS NG can be integrated with urban traffic brains, traffic control systems, computable road networks (e.g., OpenDrive, OpenStreetMap, etc.) and can also integrate with driving simulators, BIM/CIM systems, and intelligent vehicle virtual testing tools for cross-industry applications. Users can also achieve more cross-industry applications through customized services.

06

SUMOSUMO is an open-source microscopic continuous traffic flow simulation software developed by the German Aerospace Center. It comes with atraffic simulation road network editor, which allows users to interactively add roads, edit lane connectivity, handle intersection areas, and edit traffic light timing. It can also convert road networks from Vissim, OpenStreetMap, and OpenDrive through a separate conversion program. Users can specify each vehicle’s route by editing route files or use parameters to generate routes randomly. During operation, it can handle continuous traffic simulation demands for several square kilometers with tens of thousands of vehicles simultaneously, and also provides a real-time display of traffic simulation results based on OpenGL visualization.Additionally, SUMO offersconvenient C++ and Matlab interfaces, allowing for flexible co-simulation with third-party simulation programs. SUMO is primarily used for traffic flow, timing, and prediction simulations in the traffic field, and has recently begun to be applied in autonomous driving simulations, providing random complex dynamic environments for autonomous driving algorithms.

07

VIRES VTDVTD (Virtual Test Drive) is a complete modular simulation toolchain for ADAS, active safety, and autonomous driving developed by the German company VIRES. VIRES was acquired by MSC Software Group in 2017. VTD currently operates on the Linux platform, covering road environment modeling, traffic scene modeling, weather and environment simulation, simple and physically realistic sensor simulation, scene simulation management, and high-precision real-time rendering. It supports the entire development process from SIL to HIL and VIL, and its open modular framework allows for easy co-simulation with third-party tools and plugins. VIRES is also a major contributor to the widely used autonomous driving simulation open formats OpenDrive, OpenCRG, and OpenScenario, and VTD’s functionality and storage rely on these open formats.VTD’s simulation process mainly consists of three steps: road network construction, dynamic scene configuration, and simulation execution.1) VTD provides a graphical interactive road network editor (Road Network Editor, ROD) that allows users to build complex road simulation environments with multiple types of lanes while synchronously generating high-precision OpenDrive maps.2) In establishing dynamic scenes, VTD provides a graphical interactive scene editor (Scenario Editor), allowing users to add traffic bodies with user-defined behavior controls based on OpenDrive, or traffic flows running continuously in a specific area.3) Whether SIL or HIL, real-time or non-real-time simulation, single-machine or high-performance computing environments, VTD provides corresponding solutions. VTD can simulate real-time high-quality light and shadow effects, road surface reflections, vehicle rendering, rain and snow weather rendering, sensor imaging rendering, and headlight visual effects during operation.

08

rFpro

rFpro is a UK company founded in 2008, initially as an internal track reconstruction and simulation project for an F1 team, which set high requirements for simulation speed, real-time performance, and accuracy from the start. rFPro uses high-precision phase-shifting laser scanning data to scan surfaces and shoulders, producing high-resolution digital models of road surfaces with a resolution of 1cm, while using TOF laser scanning to scan roadside streets and scenes. This method provides highly matched virtual scenarios for dynamics simulation, ADAS, and autonomous driving testing.

In dynamic scene simulation, rFpro can connect to SUMO or Vissim to generate continuous traffic flows to fill the entire scene, and can also co-simulate with CarMaker to provide more realistic sensor and surface inputs for CarMaker’s test scenes. rFpro also offers a physics-based lighting and weather system, effectively simulating changes in daylight and weather conditions such as rain and fog.

09

CognataCognata is an Israeli autonomous driving simulation startup founded in 2016, which completed a $18.5 million Series B financing at the end of 2018. Cognata combines artificial intelligence, deep learning, and computer vision to recreate cities on its 3D simulation platform, providing customers with various simulated real-world test driving scenarios.Cognata’s technology is mainly divided into three aspects:In terms of static environments, Cognata’s TrueLife3DMesh engine uses computer vision and deep learning algorithms to automatically generate virtual simulation environments, including buildings, roads, lane markings, and traffic signs based on maps and satellite images.In dynamic simulation, Cognata establishes precise and scalable traffic simulation models and weather lighting models based on historical street traffic data, simulating various vehicles and pedestrians in real environments. The entire virtual simulation engine combines static and dynamic simulation models, simulating the interactions between sensors and various changes in the simulated environment, providing a complete feedback loop for the autonomous driving systems to be tested.Cognata’s simulation technology is supported by NVIDIA DGX Station, and in March 2019, Cognata announced a partnership with NVIDIA to leverage its powerful computing capabilities to simulate multiple virtual vehicles for large-scale testing in virtual environments.

10

RightHookRightHook is a startup based in California, USA, providing simulation solutions for the autonomous driving industry. RightHook offers a complete toolchain, including Right World, Right World HD, and Right World HIL. Right World provides a process for automatically reconstructing virtual scenes with rich details from high-precision maps while offering an easy-to-use testing case creation process. After case creation, organic expansion of the case can be achieved through AI algorithms. Right World also provides deterministic intelligent traffic simulation models that include vehicles, pedestrians, and bicycles. RightWorldHD simulates dynamics, weather, time variations, and sensors (including cameras, Lidar, Radar, IMU, and GPS), while supporting rich interfaces including NVIDIA DriveWorks, LCM, and ROS. RightWorldHIL provides support for HIL testing that combines software, algorithms, and hardware.

11

ParallelDomain

ParallelDomain is a startup founded in California in 2017. At the end of 2018, ParallelDomain received investment from Toyota. ParallelDomain is dedicated to automatically generating high-quality virtual environments, and its developed software can automatically generate the urban blocks needed for testing in a very short time.

The ParallelDomain platform uses real-world map data and can accept multiple map formats, using additional elements where map data is insufficient, relying on a procedural generation engine to automatically create virtual worlds. One notable feature is that all elements of the virtual world are adjustable and programmable, such as lane numbers, terrain types, mountain locations, road curvature, etc. ParallelDomain also provides dynamic traffic scenes for automatically generated scenes.

11

51Sim-One51Sim-One is an integrated autonomous driving simulation and testing platform independently developed by 51VR, incorporating multi-sensor simulation, traffic flow and agent simulation, perception and decision simulation, and autonomous driving behavior training. This simulation platform is based on physical modeling principles, featuring high precision and real-time simulation capabilities, used for the research, testing, and validation of autonomous driving products, enabling users to quickly accumulate autonomous driving experience, ensuring product performance safety and reliability, and improving product development speed while reducing development costs.In scene construction, users can quickly create OpenDrive-based road networks from scratch through WorldEditor, or restore road network information using real data such as point cloud data and map images. It supports importing existing OpenDrive format files for secondary editing, ultimately automatically generating the required static scenes. Users can freely configure global traffic flows, independent traffic agents, competing vehicles, pedestrians, and other elements to construct dynamic scenes, combined with simulations of lighting, weather, and other environmental factors to present a rich and varied virtual world.In terms of sensor simulation, 51Sim-One supports multi-channel simulation of general types or customized sensors to meet the testing and training needs of perception system algorithms, while also supporting various hardware-in-the-loop testing requirements. For camera simulation, 51Sim-One provides annotated image datasets such as semantic segmentation maps, depth maps, 2D/3D bounding boxes, and simulates monocular, wide-angle, fisheye, and other types of cameras. For radar simulation, it can provide raw LiDAR point cloud data, annotated point cloud data, bounding box data for identified objects, and also provides target-level millimeter-wave radar detection object data.

13

Pilot-DGaiAGaiA is an autonomous driving and ADAS development verification simulation tool developed by Peidai (Shanghai). It can restore complex roads by integrating road network databases, and can also recreate realistic driving environments using an environmental building model library. GaiA provides rich C++ and Matlab interfaces, suitable for various vehicles and systems to be tested. GaiA can generate numerous traffic participants and manually or automatically set their traffic behavior planning, even changing the aggressiveness of driving behavior. GaiA also provides high-fidelity environmental perception sensors, including millimeter-wave radar, LiDAR, cameras, etc.

14

MetamotoMetamoto, founded in 2016, is a Silicon Valley startup. Metamoto provides autonomous driving companies with “simulation as a service”, attempting to help autonomous driving companies achieve iterative development through an accelerated feedback loop. Its products mainly consist of three parts: designer, cloud platform, and analyzer. The designer can be used to add road networks, other environmental vehicles, pedestrians, and traffic lights to build a test scene, generating multiple test cases by controlling the range of various parameters. The cloud platform is responsible for scheduling hardware resources based on the test case situation, running test cases in parallel, and generating a large amount of test data. Upon completion, the analyzer can replay the simulated sensor data and various simulation information of the vehicle to debug the autonomous driving system. Metamoto supports precise simulation of various sensors including LiDAR, cameras, millimeter-wave radar, ultrasonic radar, GPS, and IMU, capable of responding differently to different materials. A notable feature of Metamoto is providing a fast method to adjust and cover the parameters of tests, supported by the cloud platform to run a large number of tests in a short time, effectively improving testing efficiency.

15

ESIPro-SivicESI Group’s sensor simulation and analysis solution Pro-SiVIC can help manufacturers in the transportation industry virtually test the operational performance of various onboard or airborne perception systems, accurately reproducing influencing factors such as lighting conditions, weather, and other road users.Pro-SiVIC can be used to establish highly realistic 3D scenes comparable to actual scenes and perform real-time interactive simulation analysis within the scenes, reducing the need for physical prototypes.Customers can quickly and accurately simulate the performance of various embedded systems under typical and extreme operating environments, providing sensor models based on various technologies, e.g., cameras, radar, LiDAR (laser scanners), ultrasonic sensors, GPS, odometers, and communication devices. For example, in the automotive industry, Pro-SiVIC provides multiple environmental catalogs representing different roads (urban roads, highways, and rural roads), traffic signs, and lane markings.

16

NVIDIADrive ConstellationNVIDIADrive Constellation is an autonomous driving simulation platform launched by NVIDIA, primarily consisting of two parts: one is a DGX server running the DriveSim software system, relying on DGX’s powerful graphical computing capabilities to realistically simulate lighting, night, and various weather changes in real environments, while the other server is equipped with a DRIVEAGX Pegasus onboard computer to run the full-stack algorithms for autonomous driving, forming a complete HIL simulation closed loop.

17

PanoSimPanoSim is a simulation software platform that integrates complex vehicle dynamics models, 3D driving environment models, vehicle traffic models, onboard environmental sensor models (cameras and radar), wireless communication models, GPS and digital map models, automatic generation of Matlab/Simulink simulation environments, graphics and animation post-processing tools, etc. It is based on physical modeling and a numerical simulation principle that balances accuracy and efficiency, realistically simulating various environments and conditions for automotive driving. It combines geometric modeling and physical modeling to establish high-precision camera, radar, and wireless communication models to support the R&D, testing, and validation of automotive dynamics and performance, automotive electronic control systems, intelligent driving assistance and active safety systems, environmental sensing and perception, and autonomous driving technologies and products.PanoSim includes not only complex vehicle dynamics models, chassis (braking, steering, and suspension), tires, drivers, and powertrains (engines and gearboxes) but also supports modeling and simulation analysis of various typical driving types and suspension types for large, medium, and small cars. It provides 3D digital virtual test scene modeling and editing capabilities, supporting modeling and editing of roads and road textures, lane markings, traffic signs and facilities, weather, and night scenes.

18

AAI

AAI (Automotive Artificial Intelligence) is a startup founded in Berlin in 2017. AAI builds a complex high-fidelity virtual environment based on high-precision maps, integrating traffic participants into the virtual simulation environment using artificial intelligence technology, and using driving behavior data from real-life experiences to train participant behaviors through machine learning algorithms, producing profiles for aggressive, moderate, and defensive drivers, with the goal of replicating the real world and realistically simulating all road users and environmental factors. AAI supports various sensor simulations and also provides analyzers for in-depth analysis of the data generated by the simulations.

19

AirSimAirSim is an open-source research project on drones and autonomous driving developed by Microsoft Research, built on the Unreal Engine. AirSim is implemented as a plugin for Unreal Engine, fully leveraging its capability to create highly realistic virtual environments, simulating shadows, reflections, and other real-world environments, while also providing the ability to easily generate large amounts of annotated data in virtual environments, and offering simple and convenient interfaces for drones and autonomous driving algorithms to access for extensive training. The latest version of AirSim also provides a version for the Unity engine, adding support for LiDAR.

20

CARLA

CARLA is an open-source simulator developed under the guidance of the Computer Vision Center at the Autonomous University of Barcelona, designed for the development, training, and validation of autonomous driving systems. Similar to AirSim, CARLA is also developed based on the Unreal Engine, using a server and multi-client architecture. In terms of scenes, CARLA provides open digital resources for creating scenes for autonomous driving (including city layouts, buildings, and vehicles) and several scenes built from these resources for autonomous driving testing and training. At the same time, CARLA can also use VectorZero’s road building software RoadRunner to create scenes and accompanying high-precision maps, and provides a simple map editor. CARLA supports flexible configurations of sensors and environments, supporting multiple cameras, LiDAR, GPS, and other sensors, and can adjust lighting and weather conditions. CARLA provides simple automatic behavior simulation for vehicles and pedestrians, as well as a complete set of Python interfaces for controlling vehicles, traffic lights, etc., facilitating co-simulation with autonomous driving systems to complete decision-making systems and end-to-end reinforcement learning training.

21

LGSVL SimulatorLGSVL Simulator is an open-source autonomous driving simulator developed by LG’s Silicon Valley lab based on the Unity engine. It provides integration with the open-source autonomous driving platforms Autoware and Baidu Apollo. Users can annotate in 3D scenes within Unity and export to high-precision map formats compatible with autonomous driving systems. It also provides support for sensor simulations including LiDAR, millimeter-wave radar, GPS, IMU, and cameras, synchronously outputting raw sensor results and ground truth.

22

Baidu ApolloThe Baidu Apollo simulation platform, as an important component of the Baidu Apollo platform, serves both to support the internal development and iteration of the Apollo system and to provide cloud-based decision system simulation services for developers in the Apollo ecosystem. The Apollo simulation platform is built on Baidu Cloud and Azure cloud services, allowing users to conduct simulation testing in the cloud using user-specified Apollo versions. Apollo simulation scenes can be divided into Worldsim and Logsim. Worldsim consists of pre-set roads and obstacles, serving as a simple and efficient unit test for autonomous driving vehicles, while Logsim is a scene extracted from road test data, reflecting the complex and variable obstacles and traffic conditions in real environments. The Apollo simulation platform also provides a relatively complete scene discrimination system, capable of evaluating autonomous driving algorithms from aspects such as traffic rules, dynamic behavior, and comfort.Apollo has also established a partnership with Unity, developing a realistic virtual environment simulation based on Unity, capable of providing 3D virtual environments, changes in roads, and weather. Recently, Baidu proposed a new data-driven method for end-to-end simulation of autonomous driving: Augmented Autonomous Driving Simulation (AADS). This method utilizes simulated traffic flows to enhance real-world images, creating photorealistic simulated scenes similar to those in the real world. Specifically, it is suggested to use LiDAR and camera scans of street scenes. The input data is decomposed into background, scene lighting, and foreground objects. At the same time, a new view synthesis technique is proposed to change viewpoints on static backgrounds. Foreground vehicles are equipped with computer-generated 3D models. Through accurately estimated outdoor lighting, 3D vehicle models, computer-generated pedestrians, and other movable subjects can be repositioned and rendered back to the background image to create realistic street scene images. Additionally, simulating traffic flows, synthesizing object placement and movement, and capturing real-world vehicle trajectories that look natural while capturing the complexity and diversity of real-world scenes.

23

Waymo Carcraftrepresents the world-leading level of Waymo’s autonomous vehicles, a core secret being its Carcraft simulator, which is key to Waymo’s autonomous vehicles being able to drive billions of miles each year. At the beginning of Carcraft’s development, this system was only used visually to replay the situation of roadside vehicles on the road; it later played an increasingly important role. Carcraft can test each new software version using replay data from real-world driving to verify algorithm improvements, discover new issues, and also build entirely new virtual scenes for testing. Every day, 25,000 virtual Waymo autonomous vehicles drive over eight million miles in the simulator to reinforce existing autonomous driving skills and test new skills. The greatest advantage of simulation testing is the ability to quickly repeat tests of important scenarios that rarely occur in real life, such as five-way intersections and merging roundabouts. The simulator allows the autonomous driving system many opportunities to practice these singular scenarios to master the corresponding skills. Furthermore, in the simulator, specific participants or traffic signals in a particular test scenario can be altered, adding additional pedestrians, thereby constructing a large number of derivative scenarios and allowing for more thorough testing of autonomous driving algorithms.

24

Tencent TAD Sim Simulation PlatformThe Tencent autonomous driving virtual simulation platform TAD Sim was specifically designed and developed for autonomous driving testing and verification, differentiating it from traditional simulation systems, built-in centimeter-level high-precision maps, constructing a digital twin system containing dynamic and static elements, testing the completeness of autonomous driving algorithms with ever-changing scenarios.

Previous Recommendations

• The first unmanned passenger car operation pilot in China opened in Beijing

• The first batch in the city! Intelligent connected passenger buses approved to be on the road in the Economic Development Zone

• Beijing releases the first standard system for high-level autonomous driving demonstration zones

• Beijing’s unmanned demonstration operation enters the commercialization pilot stage

• Kong Lei from Beijing Economic Development Zone: Will promote the construction of autonomous driving demonstration zone 3.0 stage

• Beijing’s high-level autonomous driving demonstration zone releases its first annual development report, contributing “Beijing experience” to the industry

• The first data classification and grading white paper for autonomous driving demonstration zones in China was released in Beijing

• The construction tasks for Beijing’s high-level autonomous driving demonstration zone 3.0 stage have been fully launched

• The world’s first open-source intelligent connected roadside unit operating system was released in Beijing

• Academician Zhang Yaqin from the Chinese Academy of Engineering: We are making autonomous driving a reality

• Autonomous Driving Weekly 9.12-9.18 | The first batch of intelligent transportation pilot application project list announced; Cruise will launch unmanned taxis in two US cities

• August financing summary: Continuing to focus on key technologies in the industry chain

• August summary of policies and standards related to the autonomous driving industry