ISO Level 5 is the highest level of automated driving defined by the Society of Automotive Engineers (SAE). It refers to a fully autonomous system that can drive the vehicle in all conditions that a human driver could, without any expectation that a human will respond to a request to intervene.
What are the Different Levels of Vehicle Automation?
The SAE has defined 6 levels of driving automation, from Level 0 (no automation) to Level 5 (full automation):
- Level 0 – No Automation: The human driver performs all driving tasks.
- Level 1 – Driver Assistance: The vehicle incorporates driver assist technologies like cruise control or lane centering, but the human driver monitors the driving environment and performs all other driving tasks.
- Level 2 – Partial Automation: The vehicle can control both steering and acceleration/deceleration in certain conditions, but the human driver must continue monitoring driving conditions and be prepared to take back control at any time.
- Level 3 – Conditional Automation: The vehicle can handle most driving functions and monitor the driving environment, but the human driver is expected to be responsive to requests to intervene.
- Level 4 – High Automation: The vehicle can handle all driving functions in certain conditions. No human intervention is expected, but the automated system will request human takeover when conditions demand it.
- Level 5 – Full Automation: The vehicle can perform all driving functions in all conditions. No human intervention is ever required.
Key Characteristics of Level 5 Automation
Level 5 automation has three key distinguishing characteristics:
- Full autonomy in all conditions – A Level 5 automated vehicle must be capable of performing all driving functions and monitoring roadway conditions well enough to operate on any type of road or environmental condition that a human driver could navigate.
- Zero driver intervention – The automated system is designed to complete an entire trip without any intervention required by a human driver. The human occupants are just passengers and need never be involved in driving.
- No geographic restrictions – The Level 5 system works in all geographic areas that an experienced human driver could operate, including cities, highways, rural roads, residential areas, etc. There are no restrictions on where these vehicles can operate.
Capabilities of Level 5 Automated Driving Systems
For a vehicle to achieve Level 5 automation, its self-driving system must demonstrate the following core capabilities:
- Perception – The ability to accurately detect and classify all objects around the vehicle using sensors like cameras, radar, and lidar.
- Localization – Pinpointing the vehicle’s location on a detailed map, tracking position precisely over time even when GPS is unavailable.
- Path planning – Determining a safe, efficient, and comfortable route to the destination while following rules of the road.
- Motion control – Skillfully controlling steering, acceleration, braking and signals to execute the planned trajectory and maneuvers.
- System health management – Monitoring the status of sensors and other critical systems, detecting malfunctions or degradations, and responding appropriately.
- Operational design domain monitoring – Recognizing when environmental, system or other conditions exceed the vehicle’s abilities and achieving a minimal risk state, such as coming to a safe stop.
These capabilities must be robust enough to handle the full diversity of driving scenarios the vehicle may encounter on public roads. The automated system can never be designed to hand control back to a human driver.
Benefits of Achieving Level 5 Automation
Fully autonomous Level 5 vehicles have the potential to transform transportation and provide enormous benefits to society, including:
- Increased safety – By eliminating human errors and unpredictable behavior, autonomous vehicles are expected to significantly reduce traffic accidents and fatalities.
- Increased mobility – People unable or unwilling to drive, like the elderly, disabled or very young, could gain independent mobility through automated transportation.
- Productivity gains – Passengers will be free to work, rest, or engage in leisure activities since they don’t have to drive the vehicle.
- Efficient parking and refueling – Autonomous vehicles can independently find parking spots or gas stations when not in use.
- Potential cost savings – Automating taxi and shared mobility fleets could make transportation cheaper and more convenient in some cases.
- Environmental benefits – Smoother traffic flow and ridesourcing lead to less congestion; electric, connected AVs enable advances in sustainability.
In addition to benefiting travelers directly, autonomous vehicles could also transform urban design and infrastructure by enabling narrower lanes, reduced parking, dynamic road pricing, and more coordinated traffic management.
Challenges to Achieving Level 5
Despite substantial progress in recent years, a number of technical, regulatory, and adoption challenges remain to be solved before fully autonomous Level 5 vehicles become a widespread reality:
- Corner cases – Human drivers can draw on lifetime experience to handle unexpected situations. Ensuring AVs respond safely in challenging, unplanned scenarios requires both massive amounts of testing and advanced AI capabilities.
- Sensing capabilities – Enabling 360 degree perception in all conditions, with redundancy and fail-safes, is extremely difficult. Fusing data from different sensor types into a coherent understanding of the environment remains an active research problem.
- System validation – Demonstrating extremely high reliability and safety will require combined real-world testing and simulation at massive scale across diverse conditions. Safety standards and validation methods are still evolving.
- Security – Preventing hacking and other cybersecurity threats is critical for safe AV operation and consumer acceptance. Comprehensive solutions to secure in-vehicle systems and external connections are still in development.
- Regulatory approval – Certifying fully autonomous vehicles is a complex challenge, involving vehicle testing standards, safety thresholds, liability questions, and ethical considerations around programmed decision making.
- Infrastructure – Optimizing roads and mobility systems for AVs will involve upgrades like standardized lane markings, traffic controls, and vehicle-to-infrastructure communication protocols.
- User adoption – Public skepticism remains high around using fully autonomous vehicles. In addition to improving trust through real-world experience, users may require training to properly interact with and supervision Level 5 vehicles.
Overcoming these challenges will require ongoing research, innovation, extensive testing, and collaboration between industry, government, and academia over many years. The road to higher levels of autonomy promises to be gradual rather than abrupt.
When Will Level 5 Autonomy Become a Reality?
Predictions on the timeline for availability and adoption of fully autonomous Level 5 vehicles vary widely. Some experts believe limited applications could emerge within 5-10 years, while widespread adoption on public roads may take 10-20 years. Others argue we are still decades away from Level 5 technology being safe and mature enough for broad commercial use.
The pace of progress will depend on many factors, including:
- – Advances in artificial intelligence and machine learning
- – Sensor, processor and computing capabilities
- – New validation frameworks and simulation capabilities
- – Regulatory standards for safety and security
- – Public acceptance and trust
- – Development of connected infrastructure and mobility ecosystems
- – Industry investment and rate of accumulation of real-world experience
Rather than a binary switch between today’s vehicles and fully autonomous ones, automation will likely continue to grow incrementally across more and more driving scenarios. For example, Level 4 vehicles limited to certain conditions may be an intermediate step. Some applications like low-speed shuttles, delivery robots, or driverless trucks confined to highways may also emerge earlier than general urban ridesharing.
Conclusion
Level 5 represents the ultimate aspiration of fully autonomous driving – to achieve a transport system smarter, safer and more widely accessible than human-driven vehicles. While the path to this goal remains challenging, the potential benefits make advancing AV technology a high priority for the automotive industry, researchers and policymakers.
With sustained effort on technical development and adoption challenges, Level 5 automation could transform mobility and society. But its deployment will require rigorous validation and responsible rollout to maximize the benefits and earn public trust. That means taking the time necessary to get the technology right.
