Update: 2019-09-29 16:11 Source: LUFTMY
In order to drive better than human beings, self-driving cars must first be safer than human beings. Establishing reliable visual functions for self-driving vehicles has become a major obstacle to development. By combining various sensors, developers have been able to create a detection system that can "see" the environment of a vehicle, even better than human vision.
The key to the system is diversity, with different types of sensors, which can verify whether the content of the car is correct. The three main sensors for driving vehicles are camera, radar and lidar. They work together to provide the vehicle with visual effects of the surrounding environment, and help it detect the speed and distance of nearby objects and their three-dimensional shape.
In addition, sensors called inertial measurement units help track the acceleration and position of vehicles. In order to understand the working principle of these sensors in the automobile, and to replace and improve human driving vision, let's first enlarge the most commonly used sensor, the camera.
From photos to videos, cameras are the most accurate way to create visual images of the world, especially on autopilot.
Autopilot relies on the camera installed around the left and right sides to blend the surrounding environment 360 degrees. Some have a wide field of vision, up to 120 degrees, and a shorter range. Others focus on narrower horizons to provide long-range visual effects.
Some cars even have fish-eye lenses, which include ultra-wide-angle lenses that can panoramically display what can be parked behind the vehicle.
Although they provide accurate visual effects, cameras still have their limitations. They can distinguish the details of the surrounding environment, but the distance between these objects needs to be calculated to know their location accurately. For camera-based sensors, it is also more difficult to detect objects under low visibility conditions such as fog, rain or night.
Self-driving cars use camera data to sense their environment
Radar sensors can supplement the camera's vision when driving at night and other low visibility, and improve the detection of automobile.
Traditionally, radar is used to detect ships, aircraft and weather objects. It works by transmitting radio waves in the form of pulses. Once these waves hit an object, they return to the sensor to provide data about the velocity and position of the object.
Highly accurate radar sensing technology
Like a vehicle's camera, radar sensors usually surround the vehicle to detect objects from all angles. They can determine speed and distance, but cannot distinguish between different types of vehicles.
Although the data provided by surround acoustic radar and camera are enough to reduce the degree of autonomy, they cannot cover all situations without manual driving.
For the complete unmanned function, lidar (a sensor for measuring distance by pulsed laser) has been proved very useful.
Laser radars allow driverless cars to view their surroundings in 3D. It provides shape and depth for the surrounding car, pedestrian and road geographic areas. Moreover, like radar, it can work normally under weak light conditions.
By launching invisible lasers at amazing speeds, lidar sensors can draw detailed 3D images based on instantaneous rebound signals. These signals will form a "point cloud" representing the surrounding environment of the vehicle to enhance the security and diversity of sensor data.
The lidar sensor is visualized through the laser pulse
The visualization of lidar sensor can detect objects by laser pulse. Vehicles only need to use lidar in several key locations to be effective.
Through experiments, PM2.5 is so high in 20-level office, especially in the narrow car space, smoking a cigarette will produce a huge amount of pollutants, which not only contains nicotine, tar, but also carbon monoxide and various irritants, carcinogens.
So it is imperative to install PM2.5 sensors in the car. At present, many PM2.5 sensors and air quality sensors will be installed in the car air conditioner or air purifier. When PM2.5 in the air is detected to increase, it will be displayed on the display screen, or voice broadcast, automatically turn on the car air conditioner. Or an air purifier to give the car owner a comfortable driving environment.
The LUFTMY infrared dust particle sensor GDS06 uses the principle of optical scattering to measure the dust concentration in the detection range through the conversion of optical path and circuit. The high precision dust sensor GDS06 has the characteristics of small size, high precision, low power consumption, short response time and stable operation under high dust concentration. The infrared sensor principle PM2.5 detection module can be used in air purifier, fresh air system, air conditioning with purification function, PM2.5 detector, etc.