As the market prospects of intelligent driving and autonomous driving are heated up, more and more people enter this industry to develop various related technologies and products. Among them, LiDAR sensors began to attract the attention of many people. This sensor can be used not only in ADAS (advanced driver assistance systems) and autonomous vehicles, but also in other applications such as drones, industrial automation, mapping, and robotics.
The difference between LiDar and radar
The word Radar is the abbreviation of "Radio Detection & Ranging". The difference between it and lidar is the energy source. As its name suggests, radar is a sensor that uses radio as its energy source. It is mainly used to detect the existence of a target object and determine its distance, and sometimes the angular position of the target.
Lidar (LIDAR) is the abbreviation of Light Detection & Ranging. It is a kind of radar with laser as the radiation source and is the product of the combination of laser technology and radar technology.
LiDAR technology was first proposed in the mid-1960s and developed in the 1980s in some developed countries in Europe and the United States in order to meet the special needs of nautical charting, port and harbor surveys. It was not until the early 1990s that the technology became mature.
LiDar technology for autonomous driving
Currently, there are two main solutions for unmanned driving. One is the use of lidar solutions. The components on the roof are called LiDAR. Google, Baidu, and Uber use this kind of solution, but this kind of solution is more expensive, for example, The Velodyne radar used by Google in its self-driving car prototype is priced at $70,000; therefore, other manufacturers have adopted low-cost solutions, such as deploying millimeter-wave radars and ADAS functions in the front and rear of the car. Camera.
These two solutions have their own advantages: LiDAR, similar to the sounding organ of whale sonar or bats, is based on the principle of obtaining "the travel time of a light pulse hitting an object and reflecting back to the receiver", and then according to the speed of light. The known principle is to convert the propagation time into LiDAR according to the distance of the measured object. In other words, LiDAR has an advantage in the accuracy of ranging. Using the "millimeter wave + ADAS" vision solution, the precise identification of objects, colors, etc., can be captured more directly. In fact, the visual solution is also quite accurate in the recognition of close distances. For example, our eyes can easily recognize the depth of field of close objects. But in general, in the case of a longer distance, with LiDAR can produce better results.
Ford tested an autonomous navigation driving test at night in the Witman area of Arizona, USA.
Carrying out the night environmental road test at the Ford Arizona Proving Ground, which is "out of reach", is an important step towards the development of fully autonomous driving technology in Ford's eyes. It is also substantial progress towards the promise of "bringing the convenience of fully autonomous driving technology to consumers around the world". This test shows that even without a camera that relies on visible light to work, Ford's LiDAR sensor is powerful enough to work with onboard virtual driving software to control the vehicle to drive smoothly on winding roads.
Although this test does not revolutionize autonomous driving technology, it improves people’s understanding of autonomous driving. Although for autonomous driving technology, the combination of radar, camera and LiDAR laser ranging, and positioning and navigation sensors are the most ideal. However, experiments have proved that LiDAR sensors are fully capable of "working alone" and can work normally even without car lights.
In order to drive freely in the dark environment, Ford's self-driving test vehicle uses high-resolution three-dimensional maps-these maps have completely included roads, road signs, topography, and landforms, as well as signs, buildings, trees, and other topographical landmark information. During driving, the test vehicle emits pulses through LiDAR to accurately locate itself on the map in real-time. At the same time, data received through radar can be integrated with LiDAR sensor data to further improve the comprehensive sensing capabilities of autonomous vehicles.
In the desert road test, Ford's engineers used night vision goggles to conduct a comprehensive observation of the test vehicle inside and outside the vehicle. Night vision goggles can help them clearly observe that when the vehicle is driving, the onboard LiDAR sensor continuously emits grid-like infrared laser rays around the vehicle body. The LiDAR sensor can emit laser pulses with a frequency of 2.8 million per second to accurately scan the surrounding environment.
The technical solutions for autonomous driving are still being continuously improved. Some voices in the market believe that autonomous vehicles do not necessarily need LiDAR, and some say that autonomous vehicles require LiDAR and advanced cameras and radar sensors to cooperate to achieve true autonomous driving. More experiments have proved that LiDAR sensors are fully capable of "working alone". But all this shows that autonomous driving technology is getting closer and closer to us.