Shutterstock | Karsten Neglia
A key concern to automobile designers is the interaction between vehicles, people, and road infrastructure.
In this equation, people are the main element but they are also the weakest link as almost all traffic accidents occur due to human errors.
Currently, advanced driver assistance systems (ADAS) are a major focus area, with automobile producers racing to create the first feasible ‘autopilot’ systems for advanced cars that would be driverless or partially driverless.
To achieve this advancement, state-of-the-art detection systems are crucial to provide the car with an exact image of its external surroundings. The US is estimated to have 3.5 million highly automated vehicles by the year 2030, due to the rapid growth of testing infrastructure and alternative mobility solutions.
Light Detection and Ranging (LiDAR)
Light Detection and Ranging (LiDAR) is a light-based system that utilizes high-speed optical time-of-flight surveying to bounce off objects and generate a high-resolution map of the surroundings in real time.
RADAR (radio waves) and cameras (computer vision) can detect objects and measure distance, but LiDAR offers a combination of these systems with the additional ability to identify the shape of those objects with high precision. Although RADAR systems have formerly been the ‘gold standard’ in ADAS, LiDAR is currently able to compete on price, providing a lower cost solution with outstanding performance.
Detection of cars, motorcycles, bicycles, and pedestrians with ranges of up to 100 m, can be achieved at a price lower than many RADAR solutions currently available. A combination of RADAR and LiDAR in an ADAS system would offer a level of redundancy that, in combination with neural networking, high-level multi-layer mapping, GPU platforms, and developments in artificial intelligence, will make driverless vehicles a real force in the automobile sector.
Fully autonomous vehicles require redundancy to provide alternatives to the limitations of current technologies, such as cameras and RADAR. This also offers protection from system errors of one or more of the sensors, and enhances performance in all environmental conditions.
For instance, cameras would require extra sensors (e.g. RADAR or LiDAR) to ensure safe automobile operation, as they may not be effective in rain or snow, or at night.
LiDAR functions better than most traditional ADAS sensors in the accurate profiling of dynamic or static objects approaching over long distances. Also, LiDARs are less likely to provide missed or false alerts, making them much safer.
Integrating a LIDAR system with a RADAR system and/or a camera vision can help to create a comprehensive ADAS system.
In the past, scanning LiDAR systems were known for their poor aesthetics and large, bulky size. This has limited their potential for utilization in automobiles where both aesthetics and functionality of the finished vehicle are important for consumers.
However LeddarTech has developed concept systems, which consist of 360° view micro-LiDAR systems located in the headlights and taillights of a vehicle. This location not only eliminates reduction in aesthetics caused by the sensors, but also prevents sensor blind spots.
As LiDAR offers various advantages for ADAS-equipped vehicles, it is extensively used in the automotive sector.
Automotive manufacturers who are leading in the LiDAR marketspace include:
- Bosch’s ‘highway pilot system’ - a semi-autonomous system capable of assuming all driving duties on open highways
- Valeo launched the Scala laser system and solid state LiDAR system in May 2016 as part of its ‘Intuitive driving initiative’
- Continental introduced an innovative driverless demo in a Chrysler 300cc in late 2015
Existing technology that uses only RADAR and computer vision (cameras) lack redunancy and can struggle when exposed to adverse weather such as snow and heavy rain. Shutterstock | Blue Island
LeddarTech is a spin-off company established in 2007. It is based in Quebec City, Canada, and has been developing optimized LiDAR technology since 2003.
The company has LiDAR expertise with 54 patents and more than 20 million hours of LiDAR operations in an outdoor environment. Their automotive work includes projects with top companies such as Morpho, Valeo, and Transcore. LeddarTech has assisted Valeo since 2011 and they recently launched a solid state LiDAR sensor that uses LeddarTech technology.
The flagship sensor of Leddar technology is a multi-segment solid-state LiDAR sensor, which is extensively reported to have one of the best specifications available commercially. LeddarTech offers a number of different sensor configurations, which can realize ranges of up to 250 m, an estimated 32,000 measurement points per frame, and an angular resolution of 0.25° based on their newest published roadmap.
A single Leddar sensor can achieve this remarkable depth of measurement and resolution beyond a 140° panorama. Using a number of sensors in a single system, a fully integrated 360° ‘view’ can be realized for ADAS automobiles and high-resolution mapping.
What Makes Leddar Technology Stand Out?
LeddarTech’s LiDAR technology offers many distinct benefits compared to its competitor’s technology. These include:
- Increased maximum range
- Increased sensitivity
- Improved field of view
- Reduced power consumption
- Excellent performance in unfavorable conditions, including low or high ambient light (using no filters, even in direct sunlight), and poor weather conditions (rain, snow, and fog)
- When sensors overlap and share some of the same field of view, there is no low performance problems
Processing signals is another important factor in LiDAR systems. LeddarTech has also delivered many benefits in this area. These include:
- Enhancement of the dynamic range to 80 - 100 db (60 db more than similar products)
- Detection of poor signal quality and environmental changes
- Discrimination of multiple objects in the same photo detector segment down to 15 cm enabling object classification, discrimination, and tracking so the course of objects in the field of view can be predicted and risk of collision determined
- Measurement of reflective and natural targets down to 0 m
- There are no range ambiguities
Why Choose LiDAR over Radar and Camera Systems?
It is agreed that for the greatest safety in ADAS systems it is better to implement multiple detection systems, however LeddarTech believe that LiDAR offers several advantages over both camera vision systems and RADAR.
LiDAR has demonstrated optimized range and accuracy, and is also more powerful in poor weather and low light conditions compared to camera systems. RADAR provides several false positives and functions poorly in tunnel-like environments because the RADAR signals echo from the walls. Therefore, LiDAR is considered to be more reliable and possess a higher resolution than RADAR with improved object discrimination.
The Rise of LiDAR
LiDAR is set to become a vital part of ADAS and autonomous driving systems. LeddarTech’s solid-state LiDAR systems are the ideal choice for automotive applications, providing a superior record of robust operations and offering excellent performance for the lowest size and price.
Predicted to be launched in 2018-20, advanced LeddarCore ASSPs (Advanced Solid-State Photonics) will offer economical, high density 3D point cloud LiDARs and provide support for both the flash and beam steering LiDAR needed for autonomous driving systems.
A close collaboration can be anticipated between automotive suppliers and complementary technology partners to meet the demand for improved LiDAR systems from manufacturers. It is anticipated that LiDAR technology will be combined into currently available camera and RADAR systems to offer complete redundancy in the case of a systems error, and also to offer accurate coverage in unfavorable weather conditions, where optical and RADAR systems fail.
This information has been sourced, reviewed and adapted from materials provided by LeddarTech Inc.
For more information on this source, please visit LeddarTech Inc.