TRANSPORTATION AND LOGISTICS BRIEFING: The future of urban transportation - Nvidia's new computer for fully driverless cars - Baidu nabs another self-driving partner
Have feedback? We'd like to hear from you. Write me at: firstname.lastname@example.org .
HOW CITIES CAN BENEFIT FROM CONNECTED, AUTONOMOUS TRANSPORTATION: Cities are grappling with decaying transportation infrastructure and booming populations driven by global urbanization. Meanwhile, new, digitally-enabled transportation services like ride-hailing are increasing total miles driven in urban centers, exacerbating traffic congestion, according to a new study from the University of California Davis. Managing these growing challenges will require cities to adopt software platforms to track and manage traffic across all modes of transportation, according to a new Deloitte report .
Such cross-service transportation management platforms, or "mobility operating systems," would:
- Leverage APIs and other tools to connect traditional and alternative transportation services, including connected bus and rail services, and bike- and ride-sharing apps. This would allow the platform to provide citizens with trip planning tools that can help them get to their destination in the fastest or cheapest way possible using multiple services. It would also allow the city to manage public transit with real-time alerts, and give private transportation providers access to new data about consumers' transportation habits.
- Collect real-time data from connected infrastructure, including smart parking meters and traffic lights. This will allow cities to track and manipulate street traffic in real time to ease congestion. Eventually, that infrastructure would include vehicle-to-infrastructure communication that would communicate directly with autonomous cars to help them navigate streets and find parking.
- Integrate payment and ticketing across all modes of transportation, allowing the city to monetize the platform by turning it into a centralized payment mechanism for all transportation services. Additionally, the city could incentivize certain modes of transportation through discounts to relieve train delays or congested streets.
These capabilities could help cities improve the throughput of their transportation infrastructure - meaning they could move more people more quickly through their existing transportation networks, according to Scott Corwin, managing director of Deloitte's Future of Mobility practice, and co-author of the study . Funding and managing stakeholders' interests to build and manage such projects will be a major challenge for cities, Corwin said. However, the value proposition of monetizing the operating systems through ticketing and payments should help attract private sector funding, he noted. Additionally, the operating systems will lure private transportation services with data on where demand for their services is highest, and those services could bring millions of their existing users into the platform by joining.
NVIDIA'S NEW COMPUTER FOR LEVEL 5 AUTONOMOUS CARS: Nvidia, one of the leading players in the self-driving space, announced a new computing system, dubbed Pegasus, which the company says could power a a Level 5 autonomous car, The Verge reports . For context, Level 5 is the highest stage of autonomy when a car can drive itself in any scenario, making steering wheels and pedals unnecessary.
The Pegasus platform builds on top of Nvidia's existing Drive PX computing systems that are already used in many self-driving vehicle tests around the world. Nvidia's core graphics processing chips are popular for self-driving applications because of their ability to process large volumes of visual data very quickly. The company has partnerships with a number of automakers using its technology for their self-driving car tests, including Ford, Volvo, BMW, and Toyota. It also recently announced a joint test with logistics giant DHL, which will test self-driving delivery trucks equipped with Nvidia's Drive PX 2 computing system.The new Pegasus platform has 10X the processing power of the Drive PX 2 platform, Nvidia's previous self-driving system . Like its predecessors, Pegasus ingests all of the data that self-driving cars generate from their camera, radar, and LiDAR sensors and quickly processes that data to create a map of the vehicle's surroundings. Pegasus will be available to automakers to use in self-driving tests in the second half of 2018.
Nvidia is trying to get ahead of the market with its system for Level 5 autonomy. Audi just showed off its new 2019 A8, the first vehicle for market with Level 3 autonomy, and the automaker still needs to get approval from regulators around the world to enable the system for use on public roads. Still, many players developing self-driving vehicles, including Waymo and Uber, are aiming to reach Level 5 autonomy and remove the need for a driver eventually. Waymo has already partnered with Nvidia rival Intel to power Level 5 autonomous cars. Additionally, California just cleared the way for the many companies testing self-driving vehicles in the state to move ahead with Level 5 autonomy test trials. Removing the driver is critical to enable the autonomous mobility services that Uber, Waymo, Lyft, and many automakers plan to launch with self-driving cars. So these companies will likely be gearing up in the next year or so to move forward with testing Level 5 technology.
BAIDU PARTNERS WITH CHINESE AUTOMAKER: Chinese search giant Baidu has partnered with automaker BAIC Group to mass produce self-driving cars. The companies plan to bring a Level 3 self-driving car to market by 2019, and to introduce a Level 4 self-driving vehicle by 2021. Level 3 autonomy is when an operator is still present and will intervene when necessary, while Level 4 is when a vehicle is fully autonomous and no driver intervention is required.
BAIC is one of many automakers using Baidu's open-sourced Apollo self-driving platform. Baidu has been working on self-driving technologies since 2014. In July, the search giant open-sourced the AI and cloud services behind its self-driving technologies. About seventy partners, most of which are automakers, now use the open platform. However, while many automakers built Baidu's software into their own systems, BAIC is working with the search giant to bring the cars to market.
The companies plan to build fleets of self-driving taxis, which puts them on a collision course with ride-hailing giant Didi Chuxing. The ride-hailing giant has also been working on self-driving cars for the last few years and now has 400 million registered users in China. Baidu and BIAC have an uphill battle in getting to market and attracting users for their self-driving taxis. However, since 75% of the country's consumers are willing or very willing to ride in a self-driving car, the market will likely be large enough for several players to succeed.
In other news...
- Hong Kong-based logistics startup Lalamove raised $100 million in Series C funding at a valuation of $900 million, according to TechCrunch . Shunwei Capital led the funding round, and was joined by Mindworks Ventures and Xianghe Capital. Lalamove provides on-demand transportation and logistics services like trucking and warehousing to businesses, similar to how Uber provides on-demand rides for consumers. The company will use the new funding to boost its share of the Chinese logistics market, which CEO Shing Chow estimates is currently worth $1.7 trillion per year.
- Tesla has issued a voluntary recall for about 11,000 of its Model X crossovers over fears that a cable powering the adjusting second row seats may not allow the seats to lock into position when they fold up from the flat position, according to TechCrunch . This affects only about 18% of the 58,781 of Model X vehicles that the company has delivered since the end of 2015. The automaker says that while it believes only about 3% of the 11,000 vehicles are affected by this issue, they wanted to be extra cautious.
- Ford invested an undisclosed sum in mobility startup Autonomic, according to The Verge . The investment will allow the automaker to build out urban mobility services that exclusively use Ford vehicles, such as the company's Chariot ride-sharing program. In addition, cities and other companies will be able to link to these future services in order to provide consumers with as many mobility options as possible.