Real-World Challenges Impeding Autonomous Vehicle Operation on Public Roads

Leading tech giants and microchip manufacturers and small-time AV ventures backed by investments are persevering towards deploying AVs on public roads with different approaches towards developing full self-driving (FSD) capability in vehicles. However, there are several real-world challenges around road safety, connectivity, traffic regulations, laws and technology costs that hinder the adoption of AVs with continuously shifting timelines for commercialization.

In addition, the lack of dedicated infrastructure required for AV operation, complexities with collision-related liabilities and ownership, and other variables pertaining to the operation of AVs alongside non-AVs will delay the pace of deployment, especially for cars in the consumer market. To understand the underlying complexities, let’s dive into a few of these issues.

The quest for next-generation technology

Level 3 autonomy in the passenger vehicles market is at a very early stage of growth, with only a few instances of commercially available vehicles operating on the road at a slower speed. With Level 3 and above self-driving capability, the liability starts shifting from the driver to the AV system in the vehicle. In the case of incidents and accidents, stakeholder liabilities are very difficult to determine for insurance companies and for DoTs to regulate multiple stakeholders, which include technology participants, vendors, mobility service providers, OEMs, and consumers.

The United Nations Economic Commission for Europe (UNECE) recently announced the extension of automated driving in certain traffic environments from the earlier limit of 60 km/h to up to 130 km/h. However, other regulatory boards across key regions of the world are yet to play catch-up with mandates restricting the commercialization of AV technology in the respective region.

The other big challenge associated with the commercialization of AV in the consumer space is the cost associated with the array of sensors used in the vehicle. The LIDAR, a sensor that gives an autonomous vehicle its “eyes,” can sometimes cost more than twice the price of the car itself, limiting automotive companies’ capability to deploy the technology across all their vehicle offerings.

Tesla is one such OEM that ultimately chose to remove LIDAR from the AV sensor mix and completely rely on cameras. However, with increasing interest in AVs, sensor technology is continuously evolving, leading to a significant development of next-generation perception sensors such as solid-state LIDARs, 4D radars, and 4D cameras with substantial reductions in cost and as cheaper alternatives to long-range LIDAR technology.

Addressing real-world challenges

Other AV use cases, such as robo-taxis and public transit vehicles, are much closer to commercialization. There are several instances of robo-taxis operating in the testing phase or operating at limited capacity in parts of the United States, Europe, and China. However, some limitations associated with connectivity and geo-fencing hinder the full-scale commercialization of these AV applications on public roads. Extensive 5G coverage across the region of operation is essential to both dispatch software updates and extend onboard sensors with real-time enhanced perception as part of the HD dynamic maps that vehicles will use for navigation.

Also, remote operation capabilities are necessary as these vehicles may occasionally face situations they cannot resolve autonomously. Enabling other advanced functionalities around Cellular Vehicle to Everything (CV2X)  implementation, such as emergency vehicle preemption and cooperative driving, would require updating the current infrastructure and collaborating with various other stakeholders. All these real-world challenges need to be addressed to ensure the safe and reliable operation of AVs on public roads.

Future outlook

Nevertheless, despite the challenges, IDC believes the AV market will continue to grow with ongoing activities in advanced driving assistance systems (ADAS) and AV software stack development. Different stakeholders, both public and private entities, will come together to solve some of the above-mentioned real-world challenges.

By 2026, 1 in 5 cars in developed regions will offer one or more Level 2 and above features due to intensifying competition amongst auto OEMs. IDC recently published the Worldwide Autonomous Vehicle Forecast, 2022–2026. In this study, IDC evaluates the maximum autonomy level for all light-duty vehicles and trucks shipped during the forecast period, leveraging the requirements as outlined in the Society of Automotive Engineers (SAE) J3016 “Levels of Driving Automation” standard.

How Capgemini works with clients to define their journey to autonomous driving systems

Capgemini, a global group of more than 350,000 engineers and scientists, is solving some of the AV-related challenges via its driving assistance offering, which includes the development and validation of new features up to the L3+ level. This implies that the company specializes in managing very large amounts of data and numerous test scenarios, with complex data processing capabilities.

Capgemini supports its clients in this era of rapid transition, developing innovative features and solutions. Its expert team bridges the gap between critical safety (including cybersecurity) and systems engineering for enhanced safety and accelerates the entire verification and validation process to facilitate the driving journey in a safe way. To complement these services, Capgemini also addresses:

• Mobility experience – in-vehicle mobility development and the vehicle communication environment
• Sustainable mobility – sustainable powertrain and energy, global environmental impact, and new mobility vehicle development
• Efficient engineering and operations – engineering process optimization, full development of urban vehicles, product engineering, efficient manufacturing, and supply chain

Explore more

Capgemini in Automotive

Get the future you want

Autonomous driving

How to overcome the 5 main technology challenges?

Driving automation systems validation

Make autonomous vehicle development and driving safe, powered by data