Driverless car, future car,
Late last year, Forbes’s Brad Templeton reported that “FCC Chairman Proposes Turning Vehicle-To-Vehicle Band To Unlicensed WiFi-Style Use.” The technical change was that the FCC has approved a proposal that would reallocate the lower 45 MHz of the 5.9 GHz band for unlicensed use ( Wi-Fi), and allocate the remaining 30 MHz to transportation- and vehicle safety-related communication services. The subtext was that after two decades, the FCC grew tired of waiting for widespread usage of this band for its initial intended purpose. Meanwhile, 5G can offer many of the same benefits and is in heavy investment mode by the telecommunications industry. Predictably, the traditional transportation community was displeased, and cited the current sunk costs as well as impact on safety.
In deeper technical venues (SAE Edge), I have made the argument about the dangers of relying on V2X communication for safety applications for systems such as Autonomous Vehicles. Similarly, Brad also has an excellent thought-piece, “V2V and the challenge of cooperating technology,” which is worth reading around the technical arguments. However, the core driver for the FCC action was lack of proliferation, and the central question is: Does the structure of the public transportation sector makes it difficult (perhaps impossible) to absorb disruptive new technologies ? There seem to be three factors at work: Public vs. Private Risk Capital, Product vs. Consulting Culture, and Civil vs. Computing Technology. Let’s consider each of these.
Public vs. Private Risk Capital: In network-based business models, there is a large investment in the network which is monetized by incremental usage over time. In the case of the private sector, the monetized vehicle is built as a part of the business case for the infrastructure investment. For new technologies, there is considerable risk. Thus, there is an active capital market for funding these risks. The successes and failures for this model are well known (Facebook, Myspace, etc). With public infrastructure investments, the direct monetization model often does not exist, so there are not active risk capital markets. Thus, public agencies are rightly risk-averse.
“Expecting public agencies to provide infrastructure for emerging technologies without explicit appropriations is likely not going to work,” said Michael McGrath, expert in management consulting and author of “Autonomous Vehicles: Opportunities, Strategies and Disruptions.:
Product vs. Consulting Culture: Public transportation agencies have evolved over time to a model of a heavily service-based culture. That is, the agency itself does not do much development, but rather it provides the management and procurement, and nearly all development is done by a vast base of civil engineering consulting companies (HNTB, RS&H, etc). One can question the structure, but for traditional civil engineering projects, it seems to work reasonably well because of the depth of the civil engineering ecosystem. However, for disruptive technologies, one enters into a chicken-and-egg situation where the investment required to build and qualify capability does not exist anywhere. It is the unusual agency which has the foresight and leadership to make this investment.
Civil and Computing Technology: The large technology disruptor for the last 70 years has been the introduction of computing (e.g., mainframes for G&A, edge computing with cloud for marketing/sales, Artificial Intelligence with Internet of Things for cyber-physical systems). The majority of the people in transportation are trained civil, automotive, or mechanical engineers and all the new disruptive capabilities are centered on electrical engineering and computer science. The gulf between these disciplines is large enough that it is difficult for the players to have a common language of discourse. Of course, agencies can build competence around computing capabilities internally, but it is very difficult. Why ? Computing is impacting nearly every market and the competition for these resources is fierce. Thus, this staff is expensive, and the economic model does not support building non-billable capability.
Given the fundamental structural challenges, it is not surprising that the level of progress for V2X was not acceptable for the last 20 years. Perhaps the lesson should be to recognize these structural realities and plan appropriately for the next technology disruption.