Sets up pilot project aiming to categorize and evaluate commercial solutions
LONDON — Leti, the France-based research institute, and its micro and nano-electronics center IRT Nanoelec, have announced a pilot program with public transport and mobility services company Transdev to characterize and assess lidar sensors for their performance and safety in autonomous vehicles under real-world conditions and to create standardized benchmarks that the industry can use.
In the pilot program, Leti teams will focus on perception requirements and challenges from a lidar system perspective and evaluate a number of commercially available sensors in real-world conditions. Vehicles will be exposed to objects with varying reflectivity, such as tires and street signs, as well as environmental conditions, such as weather, available light and fog. In addition to evaluating the sensors’ performance, the project will produce a list of criteria and objective parameters by which various commercial lidar systems could be evaluated.
Marie-Sophie Masselot, a business development manager at Leti, told EE Times, “The sensors will be chosen so as to represent different lidar architectures. As one can expect, sensor behavior strongly depends on the sensor architecture (sources, detectors, scanning method). The aim of this program is to understand the behavior at system level under different real conditions and regarding a set of common representative properties.
She added that though this program is specifically for and with Transdev, “We are aiming to enlarge this offer, opening it to other partners, to better understand the real advantages and limitations of current sensors, and to feed roadmaps for future generations of sensors. With this offer we can target the whole automotive value-chain, from component providers up to car makers.”
Masselot said the aim is to be enable developers of autonomous vehicles to better evaluate the different types of lidar sensors available. “The idea is to identify the representative benchmarks and methodologies to characterize sensors under real-world environment conditions," she said. "Today, there’s a lack of standardization on sensor benchmarks for automotive applications. For sure, the autonomous vehicle of tomorrow will need to ensure safe perception, not only based on robust algorithms and powerful automotive grade processors, but also based on sensors fulfilling a set of standardized tests. It is not required to have sensors good enough under every real-world condition on autonomous vehicles. However, it is mandatory to perfectly understand the advantage and limitation of each sensor used in the vehicle to ensure perception under kwon conditions.”
“This project will build on Leti's sensor-fusion knowhow and sensor development expertise to strengthen Transdev’s testing and evaluation of sensors for its vehicles," said Leti CEO Emmanuel Sabonnadière through a statement.
Transdev’s latest transportation technologies already enable fleets of autonomous vehicles for shared mobility. The perception of the environment through sensors is essential to offer the best experience in terms of comfort and operation speed guaranteeing the required level of safety and security. The company says evaluating sensor effectiveness and robustness is critical in developing Transdev’s autonomous transport system that will allow the operation of autonomous vehicles fleets in a maximum of environmental conditions safely and securely.
— Nitin Dahad is a European correspondent for EE Times.