Autonomous vehicle (AV) adoption is increasing exponentially as consumer demand for self-driving vehicles continues to skyrocket. Yet challenges remain for designers as AV architectures become more complex, and the question of safety remains a critical concern.
By Abhishek Jadhav
As autonomous vehicles (AVs) continue to develop through advancements in wireless connectivity, embedded systems and artificial intelligence, design challenges and cybersecurity issues remain major concerns. The rollout of electric vehicles and government incentives have spurred momentum toward research and development in AVs and associated vehicle-to-everything (V2X) technologies, including perception sensors, wireless systems-on-chip (SoCs) and edge-data processing.
V2X technology is a key driver for most AVs in development. V2X tech is essentially the ability for vehicles to share important data with other parts of the ecosystem through wireless sensor systems and cellular connectivity technologies. AVs equipped with various sensors, detectors and lasers can expand the scope of the AV driver’s perception, and adding cellular connectivity to these AVs helps improve driving safety and efficiency.
In such systems, perception of the surrounding vehicle environment is enabled by sensing and interpreting external information based on a range of sensors, radar and vision systems. The perception system must provide accurate data with low latency to avoid severe consequences. Autonomous driving relies on this sensor data to make critical decisions, which, if not acted upon quickly, can be fatal and disrupt the driving ecosystem. Companies are addressing this by developing a wireless SoC supporting the latest cellular connectivity standards to reduce the latency of data sharing.
Design challenges in the V2X ecosystem
Developments in computing technologies, such as sensors equipped with machine learning (ML), computer vision and hardware acceleration, have changed the market for autonomous driving. These technologies have enabled the automation of decision-making and control of vehicles using data gathered from multiple sensors. Using communication mechanisms like C-V2X and 5G further enhances the capabilities of autonomous-driving systems, enabling better coordination and communication between vehicles and infrastructure.
The successful implementation of an autonomous-driving system relies on the ability to make reliable and prompt decisions based on the data collected. Additionally, the reliability and accuracy of the data obtained from these technologies must be continuously monitored and evaluated to ensure the optimal functioning of the autonomous-driving system. So developing and maintaining the V2X system is a complex task requiring significant technical expertise.
For the V2X system to operate efficiently, it must connect with multiple components through wireless channels that rely on new and untested technologies. To develop a successful V2X system, wireless-communication protocols, ML algorithms and sensors must be combined and prioritized within a single structure. The integration of these components poses several technical challenges and requires a multi-disciplinary approach.
The V2X system must operate within a broader system that includes many external factors, such as moving vehicles, which behave unpredictably and are subject to changing signal conditions. The V2X system must be designed to account for these external factors, as cellular-signal fluctuations can intermittently affect communication.
Inherently, V2X communication is vulnerable to hackers who could falsify information about connected vehicles. Because the vehicles sometimes move at high speeds, they are often connected with other components briefly, making it difficult to implement security features. Managing the security of V2X networks is a formidable task. The system must be strong and capable of detecting and responding to attacks to safeguard the public’s trust in autonomous driving.
Prioritization, buffering and queuing data techniques from critical security sectors should be paramount in the V2X network. The data received from different components of the V2X system should be prioritized to prevent any collateral damage in the network. The system should also be designed to be compatible with emerging technologies. This can be done by placing interlayer proxies at various points in the communication stack. This allows for easy configuration of the intermediate layers when transferring security features to new platforms.