Velodyne LiDAR is striving to make the world a safer place with the integration of our LiDAR units into autonomous vehicle technology. However, there are a lot of questions floating around about the differences between a vehicle that is truly autonomous, verses containing ADAS technology, and whether LiDAR truly is the best option compared to other sensor technologies.
ADAS vs. Autonomy
Advanced Driver Assistance Systems, also known as ADAS, are systems designed to assist the driver during the driving process with applications like lane-keeping assist, adaptive cruise control, emergency breaking, parking assist and blind spot monitoring. These systems are programmed to do specific things such as apply brake pressure when you come to close to an object, or in some cases even move your car out of a lane to avoid another vehicle in your blind spot.
The Merriam-Webster Dictionary defines Autonomy as “freedom from external control or influence”. When applied to cars, this means a system that can operate the vehicle without any interaction with the driver. Such a system usually includes multiple cameras and sensors programmed to work together to drive the car on a pre-determined route, make adjustments to avoid obstacles and pedestrians and obey all traffic laws.
How LiDAR Works
LiDAR, which stands for Light Detection and Ranging, uses laser pulses to take measurements and generate a 3D map of an environment. After the unit sends out a laser pulse, a sensor on the instrument measures the amount of time it takes for the pulse to bounce back. As light moves at a constant speed, the LiDAR unit is able to accurately calculate the distance between itself and the target.
LiDAR vs. the Competition
LiDAR’s advantage over other sensors lies in its comprehensive data collection. To achieve full autonomy, you need an accurate and abundant amount of data. LiDAR carries more information from each data point than other sensors, including x, y, and z coordinates, time, and reflectivity (the amount of reflected light or radiation produced by an object). Most license plates, street signs, and even street line paint have retro-reflective surfaces, which provide a larger laser return signal. The LiDAR unit can easily identify if a return laser is significantly larger, allowing for the sensor to identify street signs and lane markings more quickly.
A camera could be used to accomplish some of the same tasks as a LiDAR unit, but would be blinded by bright or direct sunlight, have limited visibility at night and couldn’t measure distances to capture depth perception, dimensions, or other reliable data needed for the vehicle software to make navigational decisions.
It is also often asked if Radar could be used in lieu of a LiDAR sensor. One advantage radar has is the ability to see through objects, giving it a slight edge over LiDAR in bad weather. While LiDAR units are able to send laser light through the gaps between raindrops and snowflakes, it would collect less data during times of heavy rain or snow, as LiDAR cannot penetrate severe weather as effectively as radar.
However, Radar doesn’t have the resolution to detect small objects or multiple objects moving at fast speeds, while LiDAR produces a high-resolution image and can clearly distinguish between objects. The longer the range of the LiDAR unit’s sensors, the more data can be gathered to identify objects and allow the autonomous car to react to them in a timely manner.
Velodyne’s LiDAR units, such as the HDL-32E and the VLP-32 Ultra Puck, are already being used for research in various self-driving vehicle programs from top names in the industry like Google and Ford.
The same technology that makes LiDAR so effective when integrated into ADAS systems also makes it the best choice for full-autonomy programs. The real-time scanning, high data rate, and 200 meter range of the Velodyne VLP-32 Ultra Puck makes it the optimal sensor for any autonomy program.