As the COVID-19 epidemic hasten the house-bound economy, contactless delivery service has become the primary choice of Chinese people. Autonomous vehicles (AVs) have witnessed the upsurge in new applications. Many automakers in China are dipping their toes in applying autonomous vehicles in home delivery, goods distribution, logistics, and urban disinfection operations. In global market, Cruise, Waymo, and Pony.ai already have their self-driving cars in delivering fresh food, takeaways, and medical items.
This is only a miniature of autonomous vehicles in its development history. In the future, autonomous vehicles are going to redefine the global automotive industry. BCG report predicts that 12 million autonomous vehicles will be sold worldwide by 2035. Although the mass production of Level 3 autonomous vehicles cannot be achieved in a short term, the commercialization of Level 3-4 technology under certain regions and limited scenarios is accelerating.
According to McKinsey market research, the three most important capabilities in autonomous driving (AD) value chain in the future are the AV software development, hardware production, and AV system integration and validation and whole-car integration. These three are also the biggest technical bottleneck to realize differentiated competition. Among them, software development and AV system integration and validation and whole-car integration, are closely related to simulation.
Imagine, how can self-driving cars sense their surroundings when thick fog or snow obscures their “vision” — i.e., their cameras, radar and lidar systems? In fact, an autonomous vehicle has to validate millions of driving scenarios that interacting with its surroundings, traffic, and weather to meet the rigorous security requirements.
This is where engineering simulation comes in. With the help of simulation, autonomous vehicles can be exhaustively tested and certified and it has become one of the best choices in the field of autonomous driving. In order to gain insight into the most cutting-edge simulation technology in the industry, e-works recently interviewed Scott Stanton, Senior Director of Ansys' Worldwide Engineer Solutions. Scott talked about how to ensure the safety operation of autonomous vehicles and the role of simulation in testing and verification.
Autonomous Vehicles into the Fast Lane
BCG research suggests the market for simulation software and testing is estimated at around $10 billion over the next five years. At some point, this is benefitting from the growing AV market. Self-driving cars in its nature improve driving safety and significantly reduce traffic accidents, thus improves the efficiency of the transportation system, saving time and energy. As self-driving cars liberate their passengers from driving tasks, people will have more time on work or entertainment, which greatly improves the comfort of on-road driving.
In recent years, with the rapid development of cloud computing, 5G, Internet of Things, renewable sources and other new technologies, autonomous vehicles have seen its flourishing spring. Numerous technology giants and automakers enter the market, advancing its landing and application.
In Scott's opinion, today the question is not “Will we see fully autonomous vehicles transform multiple industries in the near future?” but “Who will be first?”, since autonomous driving technology is presenting following trends:
The decreasing cost of sensors installed in self-driving cars will reduce the production cost and accelerate the mass production and upgrade.
The rapid development and application of 5G technology promotes the birth of advanced self-driving cars and speeds up the arrival of driverless cars.
In the future, autonomous vehicles will be equipped with renewable sources. The future of automotive industry belongs to both autonomous vehicles and new energy vehicles.
Autonomous vehicles will also largely boost economy. Some analyst predicted that autonomous vehicles will grow the global economy by $7 trillion. Many countries, including China, are introducing policies to support and fund the development of key technologies of smart vehicles as well as major projects for smart transportation and smart city infrastructure.
“As companies race to solve the remaining engineering challenges and launch innovative, yet practical, autonomous vehicles, simulation becomes a competitive imperative,” said Scott.
Simulation Enables Fast Validation
Today, the race is on to achieve full autonomy. Large automotive OEMs, tier 1 suppliers, and disruptive upstarts are accelerating the development of autonomous vehicles and ADAS, aiming to get ahead of the market. The safety and comfort of self-driving cars also requires a safe and stable AV system.
As autonomous vehicles are composed of a multitude of mechanical parts, electronics, hardware and software, the process of functional safety analysis can be enormously complicated. At the same time, the application of radar, lidar, V2X, sensor fusion, and deep learning all requires a large amount of testing in different cases to make sure it’s safer than human drivers.
A report by Rand Corporation suggests that demonstrating the safety of an autonomous vehicle could take over 8 billion miles of physical road testing. According to the report, suppose a fully autonomous vehicle fleet had a true fatality rate that was 20% lower than the human driver fatality rate. It would take approximately 5 billion miles to demonstrate this difference. With a fleet of 100 autonomous vehicles test driven 24 hours a day, 365 days a year at an average speed of 25 miles per hour, this would take about 225 years, which is obviously not practical.
Scott said that, since conducting road-testing over the billions of miles required to demonstrate safety is time- and cost-prohibitive, companies have turned to simulation as the only way to successfully complete an autonomous vehicle program.
On a larger scale, simulation runs through the life cycle of research and development and use of AVs. Early from the ideation to product data collection and system upgrades, engineers are taking simulation as a toolbox to achieve security and stability. Executives are also taking it as data knowledge base to accumulate valuable intangible assets, such as design process, engineering experience and digital models. For supervision and testing organizations, simulation can be used as an analysis and testing tool to evaluate the functional safety and intelligence level.
Ansys Closed-loop and Multi-scenario Simulation for Autonomous Vehicles
Simulation now can accelerate testing in a variety of complex situations and optimize the performance of sensors and algorithms, which greatly shortens the time to market.
In terms of functional safety analysis, Scott said “to address this problem, Ansys offers the medini analyze solution family, which automates functional safety analysis and seamlessly integrates this critical activity into overall product development to ensure the consistency, traceability and efficiency.”
In terms of software development
, Ansys offers its proven SCADE family of solutions for software development and verification.By numerically modeling and controlling all code generation activities, SCADE solutions equip software engineers to meet industry safety standards and deliver high levels of performance. To support autonomous product development, SCADE solutions are designed for easy integration with third-party neural network and machine learning software.In addition to improving the reliability of software code, Ansys SCADE delivers significant improvements in development time and costs, as compared to manually based code-generation methods.
In terms of verification and validation
, Ansys offers advanced solution to simulate various transportation scenarios and different sensors. For instance, sensors are the eyes and ears of any autonomous vehicle, and thus are some of its most critical components. They are also among the most complex, tasked with gathering and processing large volumes of environmental data in real time and communicating this data to a perception algorithm. Common sensor types for autonomous vehicles include radar, lidar, cameras and ultrasound.“Testing and verifying the performance of the sensors represents a significant engineering challenge,” said Scott. For example, radar sensors are typically mounted behind the front fascia of an automobile. While perfectly controlled physical testing environments, such as anechoic chambers, can help engineers design these systems, the reality is that their radiation patterns will be skewed in real-world applications by the material properties and geometric configuration of a car’s front fascia. Radar systems must be designed to operate behind the front fascia of dozens of different types of vehicles, each with its own unique geometry and material properties. Physical building and testing is simply not practical due to the time and costs involved in each design iteration. Scott added, “Ansys offers a full suite of radar and antenna simulation solutions designed to replicate real-world performance with a high degree of fidelity. By leveraging Ansys software, engineers can predict sensor performance accurately.”
By acquiring OPTIS optical simulation to Ansys's leading portfolio in 2018, Ansys now offers solutions from structures to fluids and from acoustics and semiconductors to electromagnetics and optical. With OPTIS’ closed-loop simulation, Ansys now can help automakers simulate navigating environment, including weather, traffic, and even car crashes. Ansys VRXPERIENCE simulation platform offers a 3D virtual environment, enabling engineers simulate scenarios, traffic and vehicle dynamics run time and also create custom virtual road environments and testing scenarios. With these capabilities, automakers can complete millions of miles of virtual testing within one day.
The value of Ansys simulation solution for autonomous vehicles is that it provides a series of functionalities, such as custom virtual road environments and traffic scenarios, vehicle dynamics modeling and 3D environment modeling, physics-based multi-sensor data fusion and system level simulation, intelligent head lamp lighting simulation and other functionalities.
“With OPTIS, Ansys capabilities now span the simulation of all sensors, including lidar, cameras and radar; the multiphysics simulation of physical and electronic components; the analysis of systems functional safety; as well as the automated development of safety-certified embedded software. This functionality can be integrated into a closed-loop simulation environment that interacts with weather and traffic simulators, enabling thousands of driving scenarios to be executed virtually,” Scott added.
Besides, the latest version of Ansys Cloud provides the virtual desktop infrastructure (VDI) technology. It delivers workstation-class performance to any computer with an internet connection, which makes it easy for engineering simulation and vehicle performance data recording in cloud, and fast decision-making through algorithm optimization. We can imagine that one day engineers may use AI-based simulation in autonomous vehicles. In the future, simulation may become a key factor of a management platform for autonomous vehicles, road facilities and smart transportation.
A Simulation Tool Chain of BMW
Ansys and BMW Group are creating the industry's first holistic simulation tool chain for developing autonomous vehicle (AV) technologies. The simulation tool chain will enable highly automated and autonomous driving with the first vehicle launch expected in just two years.The new automated simulation tool chain will make efficient use of BMW´s large amount of sensor data through intelligent data analytics and the creation of scenarios according to statistical relevance and AD system sensitivity.
The multi-year agreement drives the development of BMW Group's Level 3 offering and Level 4-5 technology. The scenarios will include usual driving situations and corner cases to ensure maximum test coverage. Based on these scenarios, the tool chain will perform rigorous safety assessments of the AD systems in a high-performance virtual environment.
According to Scott, Ansys and BMW will support its adaptability and openness regarding relevant interfaces and validation approaches to accommodate and foster safety initiatives. Ansys will assume exclusive rights to the simulation tool chain technology for commercialization to a wider market.
In order to boost the development of high level autonomous vehicles, BMW and other leading automotive OEMs are leveraging Ansys' broad pervasive engineering simulation solutions and experience to open a new era. In the process, Ansys will cooperate with more automakers and technology giants, speeding up the revolution in automotive vehicles.
Scott Stanton is Senior Director of Ansys' Worldwide Engineer Solutions Team and has been with Ansys over 23years. This team is recognized as industry leaders for the following megatrends: Autonomous Vehicles (AV), Chip-Package-System (CPS), Electrification and Industrial Internet of Things (IIoT). This team's responsibility is to architect and execute engineering solutions together with our customers by creating flexible simulation environments configured to their needs.