Image Credit: metamorworks/Shutterstock.com
Autonomous vehicles (AVs), more commonly known as driverless vehicles, are arguably the next big revolution in automotive technology. But with this revolution comes a challenge: how do you ensure onboard artificial intelligence (AI) can respond in real-world traffic conditions? One solution is to capture real-world data using light detection and ranging (LiDAR).
The Challenge of AVs and Real-World Driving Conditions
It is predicted that self-driving cars will become as commonplace as cruise control over the next decade. The driverless vehicle sector industry could become worth as much as $7 trillion by 2050, representing a significant change in transportation worldwide.
However, if AVs are to become a ubiquitous sight on the roads, they must respond in the same way as a human driver to real-life, unexpected conditions.
There are many different situations and conditions in the real world. Weather conditions can vary both between seasons and in the space of an hour. At night, the world is dark, in daylight there are shadows. Narrow, foggy streets can present a challenge to even the most experienced driver.
Other vehicles on the road can sometimes cause problems, some of which may be autonomous. Pedestrians, cyclists, public transport, and a myriad of other factors also need to be considered.
If AVs are not programmed to respond in these situations, there is a real danger that the car may be involved in a fatal accident. For increased safety, the onboard AI needs to react in the same way as an experienced driver. One solution for training the AI is to use LiDAR in smart, intelligent networks.
What is LiDAR and How Could it Make Autonomous Cars Safer?
LiDAR works by emitting laser pulses outwards in various directions until the signal reaches any object within the environment, reflecting the signal and sending it back to the receiver. This is precisely the same principle as Sonar, except that LiDAR uses light rather than sound waves.
Light is 1,000,000 times faster than sound. Because of the high speed, processors need to process data points from a considerable amount of laser pulses every second. The software then produces a 3D map of the environment based on every reflection point of the laser. The real-time maps created by LiDAR software are highly accurate.
LiDAR sensors perform better than other sensor types in critical cases for traffic analysis and monitoring applications. They offer 24/7 availability due to superior perception capabilities in all lighting conditions, including night-time, intense headlight glare, and shadows. They also provide excellent angular resolution compared to radar-based systems and minimize false positives and negatives.
As driverless cars need to react in real-time to any number of unpredictable elements, the ability to gather data on the split-second scale is highly advantageous. LiDAR is uniquely placed as a technology to solve these issues, and many self-driving cars under development already these detectors.
ALP.Lab and Cepton are working in conjunction with each other to implement LiDAR-based solutions to train autonomous vehicles.
ALP.Lab and Cepton - Using LiDAR Intelligence to Increase Road Safety
Founded in 2017, ALP.Lab provides testing services for the AV industry. Cepton was founded in 2016 and provides mass-market LiDAR solutions for a variety of industries.
ALP.Lab and Cepton are working together to drive the future of safer autonomous vehicle deployment. Cepton’s HeliusTM Smart Lidar System is being installed in ALP.Lab’s testing area in Austria.
The HeliusTM system combines Cepton’s state-of-the-art Micro Motion Technology (MMT®) LiDAR technology with cutting-edge software to provide anonymized and intelligent 3D perception. The Helius system takes cloud point data from a connected sensors network, delivering real-time three-dimensional information on object locations, dimensions, and velocities.
Cepton - Helius® Smart Lidar System
Video Credit: Cepton Technologies/YouTube.com
This ability enables accurate tracking and classification of many different objects involved in a traffic situation. It also collects anonymized data, allows complete privacy of gathered data, and protects pedestrians and vehicles' anonymity while feeding back rich and accurate traffic data to ALP.Lab. This enables deep analysis of real-life conditions.
The system also offers optimal ease of integration. With a rotation-free, frictionless, and mirrorless design, the Vista-P60 sensors are durable. Featuring a rugged design, they can stand up to harsh climates and environments. This means that the product is robust and has longevity, driving down maintenance costs. The system is also scalable and easy to install, making it a truly “plug-and-play” technology.
Several of Cepton’s Vista®-P60 LiDAR sensors have been installed at the corners of intersections and roundabouts across Austria to complement other sensor types, including Radar and optical sensors. Vista-P60 sensors can detect pedestrians and vehicles at a distance of ~170 m around intersections. This makes the system a superior choice for this application.
Through this collaboration, APL.Lab and Cepton are leading the way in using LiDAR sensors for road safety.
Lidar – An Innovative Solution for the AV Industry.
Lidar technology is playing a crucial, innovative role in the future of driverless vehicles. It is serving as vehicles' eyes and using it to provide intelligence to the road infrastructure means that autonomous cars can become safer. Through research and projects such as APL.Lab and Cepton’s collaboration, AVs are becoming a reality for the broader public and will become a more common sight on our roads in years to come.
References and Further Reading
Intel.com – Accelerating the Future: The Economic Impact of the Emerging Passenger Economy. [Online] https://newsroom.intel.com/newsroom/wp-content/uploads/sites/11/2017/05/passenger-economy.pdf?cid=em-elq-26916&utm_source=elq&utm_medium=email&utm_campaign=26916&elq_cid=1494219 (Accessed 3 March 2021).
Cepton.com – Helius™. [Online] Available at: https://www.cepton.com/products/helius (Accessed 3 March 2021).
ALP.Lab. [Online] Available at: https://www.alp-lab.at/ (Accessed 3 March 2021).
Businesswire - ALP.Lab and Cepton Deploy Lidar Intelligence to Enable Smart Intersections and Road Safety in Europe. [Online] Available at: https://www.businesswire.com/news/home/20210222005304/en (Accessed 3 March 2021).
Cepton – About [Online] Available at: https://www.cepton.com/company/about (Accessed 3 March 2021).
Archer-soft.com. Lidar vs Radar: pros and cons for autonomous driving [Online] Available at: https://archer-soft.com/blog/lidar-vs-radar-pros-and-cons-autonomous-driving (Accessed 3 March 2021).