Autonomous vehicles & systems
Work with us in APPL.AI
Contact us about Autonomous Vehicles and Systems
Huge potential for AI in autonomous vehicles and systems
There are huge AI opportunities in the domains of mobility and security. Faster, lighter and more powerful sensors are improving AI-enabled autonomous vehicle systems. Excellent statistical results of current AI methods are improving the performance of autonomous vehicles in complex environments. These systems can support (and even replace) human decision-making when dealing with large datasets, analytical problems or time-critical tasks. They improve the quality of decision-making and minimise human error. In cases where humans are still better capable of performing the task, the AI system must hand over control to humans. And in doing so allowing humans to have sufficient time to adapt to the new situation.
AI in mobility has enormous potential to increase safety and efficiency. Self-organising logistics could reduce transportation kilometres (lowering costs and environmental impact). AI-supported traffic management could improve traffic safety, network efficiency and reduce traffic jams. Meanwhile, connected and autonomous vehicles could optimise traffic flow. It helps drivers increase safety, comfort and efficiency. Surveillance robots will be able to monitor areas almost continuously.
Overcoming unreliability, privacy issues and a lack of data
Autonomous vehicles and systems have enormous potential. However, several challenges need to be addressed. The world through which autonomous driving systems must navigate is extremely complex. This places significant demand on AI. Some cases have shown disturbing problems characterised by unreliable behaviour. Some of these problems occur, because AI models in autonomous driving systems are unable to cope with new situations. The AI models have little or no data. Alternatively, they are unable to provide users with the necessary transparency in the decision-making process.
Even before deploying AI, we must guarantee reliability and safety in all possible situations. We need to ensure that a human is always in control. This can be achieved by ensuring that humans can make informed decisions on everything that could result in serious harm or ethically undesirable situations. In case there is insufficient time for this, these decisions must be discussed beforehand. Privacy, data ownership and ethical issues also pose challenges in the adequate implementation of autonomous driving systems.
What does TNO offer?
- We offer a safe, open environment for shared research.
- We balance the economic benefits of using autonomous vehicles systems with ethical and legal constraints and environmental impact.
- We develop the necessary protocols for governments and automotive manufacturers to test for safety.
- We test and validate autonomous driving systems for manufacturers.
- We develop architectures and algorithms for performance/health assessments.
- We develop algorithms for connected vehicles, using our multidisciplinary domain knowledge in traffic- and vehicle-management.
- We develop algorithms and techniques that allow humans and highly autonomous systems to collaborate
- We develop verification & validation techniques for joint human-machine systems.
- We offer integral solutions by combining TNO domain knowledge from multiple and relevant disciplines (e.g. data scientists, psychologists, lawyers, environmental scientists etc.).