As we enter a period of autonomous vehicle pilot deployments it is instructive to understand how far we’ve come in making these vehicles an achievable reality, and what we still need to accomplish before autonomous fleets can address the mobility goals, we’ve set for consumer transportation and logistics. Technological innovation and adoption typically follow particular lifecycles. Think of the portable music players. Driverless mobility, comprising of on-demand mobility services and autonomous vehicles, is no exception. This post presents an analysis of the autonomous vehicles innovation lifecycle. It introduces four dimensions for assessing AV innovation performance over time. Finally, it presents six requirements that will need to be addressed before the use of autonomous vehicles can scale for consumer transportation and logistics.
Innovations,including technology innovations, advance along a lifecycle that is
characterized by four phases: Research,Growth, Scale, and Maturity. During the Research phase new technology is invented. The performance of this technology is low. For example, consider the performance exhibited by the vehicles that participated in DARPA’s first autonomous vehicle grand challenge in 2004. They were able to travel slowly for just a few miles. The Growth phase starts when new companies introduce products that improve upon the original innovation by addressing stated or perceived market requirements. Early adopters start using the product or service. During the Scale phase more companies enter the market both to satisfy the growing demand, but also to introduce new products whose innovations address additional market requirements. Adoption expands from the early to the late majority. Finally, during the Maturity phase the initial innovation and its follow-on refinements reaches a steady state and is adopted even by laggards. During this period the goal of the companies participating in the market is to prolog their established business models by introducing the smallest number of sustaining innovations that will address the market’s needs.
When we measure the performance of a successful innovation over time, we tend to see a specific pattern that often has the shape of an S-curve. Moving along the curve requires incremental innovations to address market requirements. Successfully addressing these needs leads to increased market adoption of the product or service by various groups, from the early adopters to the laggards. Sometimes an innovation that has matured is
disrupted by a new innovation that marks the beginning of a new lifecycle. For example, over the past 40 years computing has seen three different lifecycles, all of which became S-curves. The innovation of the personal computer, exemplified by the IBM PC, disrupted mainframes and minicomputers which by the mid-80s had matured, and started a new lifecycle. In the mid- to late-90s the disruptive Netscape browser ushered the era of internet computing and marked the beginning of another lifecycle. About ten years later the introduction of the disruptive iPhone marked the beginning of the mobile computing lifecycle.
Next-Generation Mobility Innovation Lifecycles
The driverless future promised by next-generation mobility is expressed today in terms of two innovations: on-demand mobility services, e.g., ride-hailing, and microtransit, and autonomous vehicles used for consumer transportation and logistics. Currently it is necessary to evaluate the performance of these innovations using two distinct lifecycles. Figure 1 shows where we are in each of these lifecycles.
Figure 1: Next-generation mobility innovation lifecycles
The autonomous vehicles innovation lifecycle we are tracking starts around 2004 with the announcement of DARPA’s first Grand Challenge. The third, and final, Grand Challenge competition was held in 2007. Google started Project Chauffeur, its autonomous vehicle project, in 2009. By the middle of the current decade, venture investors and corporations started investing in next-generation mobility resulting in a large number of startups. To date companies offering on-demand mobility services have raised globally $75B.
Over 50 OEMs and Tier 1 suppliers around the world have established automated and autonomous vehicle programs. CB Insights reported that
during first three quarters of 2018 VCs committed $4.2 billion in the sector, compared to $3 billion in 2017 and $167 million in 2014. Finally, AlixPartners reports that an additional $61B has been earmarked for AV technologies.
As we are about to enter a new decade, certain startups are breaking out and raising large financing rounds, partnerships involving large corporations and startups are accelerating, consortia are starting to form, Waymo, GM, Transdev, and Starship Robotics launched autonomous vehicle pilots and more corporations will join them in the next 12-18 months, initial vehicle orders for autonomous fleets are placed, and the importance of data and AI across the entire value chain is recognized. With all these activities, particularly the launch of the pilots, we are entering the autonomous vehicle innovation lifecycle’s Growth phase. In an earlier piece I had identified six use cases for autonomous ground vehicles: specialized vehicles operating in controlled environments, long-haul trucks, passenger shuttles, robotaxis, delivery vans, including sidewalk robots, and privately-owned vehicles. Each of these cases is at a slightly different position along the autonomous vehicle innovation lifecycle. So, while Figure 1 shows the average position along the autonomous vehicle innovation lifecycle, Figure 2 depicts the position of each use case separately.
Figure 2: The position of autonomous vehicle use cases along the innovation lifecycle
Figure 2 shows that we are further along with specialized vehicles that work in controlled environments, like mines, followed by low-speed passenger shuttles operating on relatively small geofenced areas, and sidewalk delivery robots that are part of the delivery van use case. We are starting with robotaxis and small delivery vans. Long-haul trucks, larger delivery vans, and finally privately-owned vehicles will follow in the next few years. As we start to increasingly utilize autonomous vehicles for passenger mobility and goods delivery, the two lifecycles will merge. For example, we are starting to see pilots for grocery delivery, microtransit, and ride-hailing that use autonomous vehicles.
Autonomous Vehicle Innovation Performance
Autonomous vehicle innovation performance is a complex concept. We choose to define it in terms of four dimensions: Trust, Vehicle Reliability, Business Model, and Regulation.
- Trust measures the certainty that the autonomous vehicle will transport passenger and/or goods to the intended destination
safely (no crashes) and efficiently (the vehicle will not take circuitous routes in order to avoid situations under which its systems may not perform well). Trust plays a different role depending on whether the autonomous vehicle is transporting people or goods. In the case of people transportation, trust has a stronger association with safety (no harm will be done to the passengers being transported) and incorporates notions of regulation and reliability (see below). To a lesser degree, and particularly in car-centric societies like the
American, trust may also measure the public’s willingness to cede transportation control to a robot. In the case of goods transportation, trust
is associated with the prompt delivery of undamaged cargo.
- Vehicle reliability measures the vehicle’s ability to: 1) operate correctly for the duration of each ride without requiring human intervention, implying that it will not experience disengagements, 2) perform autonomously with consistency at a level defined by one or more regulatory bodies, 3) provide rides of acceptable quality (for example, today many passengers complain that robotaxis often stop abruptly, or
moving slower than traffic). Reliability depends greatly on technology. In this case on four technology sets: vehicle technology, information technology, transportation infrastructure technology, and communication technology.
- Business model measures the operator’s ability to create a scalable business that will, over time, result in sustainable profitability. Identifying scalable business models is a complex issue. And, obviously, consumer transportation will employ different models than logistics.
For on-demand consumer transportation identifying scalable business models requires mobility services companies to consider several use cases. For example, consider the case of consumers who today use Privately-Owned Vehicles (POVs) for their daily commute from the suburbs to the city, but use on-demand ride-hailing for their inner-city transportation. In order for these consumers to use ride-hailing or ridesharing for their daily commute, the Autonomous Vehicle price/mile must be similar or lower to the cost/mile of their POVs, where
cost/mile incorporates fuel, insurance, parking and other costs associated with POVs, as well as vehicle depreciation. Consumers who use on-demand mobility services while out of town, e.g., traveling from the airport to the hotel, may have different price-points allowing the mobility services provider utilizing autonomous vehicles to charge a higher price/mile. As a third case consider the consumers who live in cities and already use ride-hailing. For them a lower price/mile afforded through the use of AVs, compared to the price/mile of human-driven vehicles, will convince them to further expand the use of such services.
- Regulation measures the extent and appropriateness of available laws and processes that govern the operation of autonomous vehicles and the fleets that use them. Regulation has to address a) the vehicle’s components that impact its autonomous operation (hardware, software), b) the data that makes that operation possible (infrastructure data, vehicle-generated data, fleet operation and management data, data coming from the various content providers participating in autonomous mobility), and c) the data that is generated by the users of the transportation and logistics services.
Figure 3 shows the performance for the autonomous vehicle lifecycle in terms of the four identified dimensions. It presents where we are today along these dimensions, and where we need to be in order to reach scale.
3: Four dimensions characterizing autonomous vehicle innovation performance
Because AV technology has improved dramatically over the past 10+ years, vehicle reliability is ahead of every other dimension, with regulation being the farthest behind. Improving trust, particularly for consumer transportation, will require various forms of education coupled with advertising campaigns. Regulation is primarily addressed through advocacy and lobbying, with a certain level of trust and reliability being prerequisite. Regulation relating to autonomous vehicles will not be an exception.
Autonomous Vehicle Safety
Surveys about consumer intent to utilize transportation that is provided by autonomous vehicles showed that passenger safety while riding in these vehicles is a concern. But besides passenger safety autonomous vehicle technology developers and regulators will need to concurrently address the safety of those outside the vehicle particularly since cities consider autonomous vehicle an important component to their Vision Zero planning.
We define safety as a function that combines Trust, Vehicle Reliability, and Regulation. Figure 3 shows results from research conducted by the Pew Center in 2017 where 30% of those surveyed cite AV safety as a primary passenger concern.
Figure 4: Pew Research survey results on autonomous vehicle safety
It appears that after the accidents that occurred early in 2018, starting with Uber’s fatal accident, the public became even more concerned about autonomous vehicle safety. Data from Deloitte’s most recent annual Global Automotive Consumer Study (Figure 5) shows that across different geographies a significant percent of surveyed consumers are concerned about AV passenger safety.
Figure 5: Deloitte consumer survey results on autonomous vehicle safety
It is necessary to distinguish between the passenger safety relating to the AV Stack, i.e., the collection of hardware and software that under a particular configuration enable the vehicle to move autonomously, and the vehicle’s crashworthiness. For example, assuming they use the same AV Stack and operating in the same urban environment, a Hummer robotaxi has higher crashworthiness than an autonomous Prius because the Hummer’s design, construction, weight, etc., provides better protection to its passenger than the Prius in the event of a typical urban vehicle crash. Therefore, the Hummer may be perceived as safer than the Prius. However, this performance says nothing about the reliability, and trustworthiness of the AV Stack used by these vehicles, particularly under the current light regulatory environment. Today the distinction between crashworthiness and AV Stack safety is at the heart of the legislative and regulatory discussions. Even autonomous low-speed shuttles lacking a steering wheel are by default illegal. Designs such as the fourth-generation Cruise autonomous vehicle, and the (retired) Waymo Firefly pod are considered non-compliant. RAND provides a detailed framework for assessing autonomous vehicle safety across the development, validation, and deployment phases of these vehicles.
Six Requirements for Growing and Scaling Autonomous Mobility
I have identified six requirements that will need to be satisfied in order to move autonomous vehicles from their current position innovation lifecycle initially to the Growth and ultimately the Scale stage. Satisfying these requirements will improve improving the performance along all four of the identified dimensions.
Creating the appropriate regulations that will provide the right controls without stymieing growth (consider the impact on Uber’s growth when London restricted the company’s operations) should become a high priority. Policy-makers must accelerate their efforts in order to catch up with the progress we have made on technology and reliability, and the corresponding progress we are making on business models.
Requirement 2 (Developing trust through testing): In order to develop trust in autonomous vehicles, each of their subsystems will require thorough testing. Many of the software subsystems are AI-based. Several of these utilize statistical machine learning and their behavior is not deterministic. As a result, their testing and validation becomes harder requiring the development of new testing approaches. This testing process is very different from the one used for the validation of the software systems found in today’s conventional vehicles.
Requirement 3 (Education): Companies and governments involved in next-generation consumer mobility must embark on various forms of education aimed at transforming the public’s anxiety to excitement about and comfort with riding in autonomous vehicles. Similar education will be necessary for truck drivers who are currently concerned that autonomous trucks will result in significant job losses even though the demand for truck drivers cannot be filled. Automated and autonomous logistics vehicles must be positioned as providing safety enhancements, increase the quality of the driving experience, and improve productivity.
Requirement 4 (Business models): Accelerate work that expands the deployment of autonomous vehicles under the appropriate business models under the use cases that are possible with the available technology. As it is shown in Figure 2 the technology has progressed to the point where we can safely operate low-speed autonomous shuttles, and sidewalk robots used for last-mile deliveries in small geofenced areas and under specific weather and traffic conditions. Broad adoption of autonomous vehicles will require scalable business models. Under such models, profits must be generated consistently while the price/mile of the service that uses autonomous vehicle is similar to or lower than the price/mile of the same service that uses human-driven service (passenger transportation or logistics) and the cost/mile of POVs. For this reason, scalable business models for autonomous logistics may emerge than for consumer transportation.
Requirement 5 (Investment): Aggressive investment in autonomous vehicle technology development, and in infrastructure (transportation, communication, electrification) development must continue even if the timeline to ROI may be longer than originally envisioned. The public is starting to realize that developing autonomous vehicles for consumer transportation or logistics is a very hard task, much harder than certain technology developers, business executives and the media had us believe. Even though we have achieved 90% of the necessary functionality, the last 10% may require 90% of the effort. Because of their magnitude, future investments may not come from VCs but from private equity investors, e.g., the Vision Fund, sovereign funds, e.g., Temasek, whose countries consider autonomous mobility a national imperative, or corporations and corporate consortia, e.g., see Honda’s investment in GM’s Cruise unit.
Requirement 6 (Big data and AI): Invest in data management and AI technologies that will be applied for capturing, storing, and analyzing, or otherwise exploiting, the big data that is generated by autonomous vehicles and the environments and infrastructures where they operate. Alongside, establish regulations for safeguarding data privacy and ensuring cybersecurity. Among other benefits such data and AI technologies will enable us to test and validate faster the performance of autonomous vehicles under a variety of scenarios, as well as assess the effectiveness of business models, including data monetization models.
(Cross-posted @ Re-Imagining Corporate Innovation with a Silicon Valley Perspective)