With the gradual improvement of material living standards, more and more advanced science and technology products have turned into people's toys. In retrospect, the toys in our hands used to be paper planes, marbles, and gas-fired ammunition that wealthy children might buy. In the twinkling of an eye 15 to 20 years later, our grown-up children and our toys It became a small computer on the wrist, or a GoPro motion camera capable of recording 4K frames at high frame rates in the palm of your hand.
In all the "plaything", drones are probably the most expensive, most do not like a class of toys. They are powered by electricity, lift through multi-rotor flights, and airborne HD cameras for a variety of uses, including aerial photography and geo-mapping. It can be said that UAV has been upgraded from a toy to a multi-purpose technology product because of its convenient use and powerful functions.
But at the same time, UAVs are always out of sorts of social news because of security issues. A few months ago, a Xinjiang drone crashed on the lawn of the White House where the United States presidents home, attracting widespread concern over the safe operation of unmanned aerial vehicles. The FAA, the US air traffic control agency, has also in the past been rigorously controlling the commercial use of drones for related issues until the first FAA-approved commercial UAVs were only a few days old (July 19, 2015) Express delivery.
UAVs want to continue to develop, in addition to battery life, functional improvements, safety is an undeniable aspect. As time goes by, in 2015, major consumer and commercial UAV makers in the world are starting to focus on Obstacle Avoidance as a product focus. Today, to tell you about the mainstream UAV currently on the market, such as Xinjiang, Parrot and so have adopted what kind of automatic obstacle avoidance system?
After visiting and interviewing, we find that there are mainly three main types of obstacle avoidance systems for electric multi-rotor UAVs: ultrasonic sensors/ultrasonic transducers, TOFs, and relatively complex composite methods consisting of multiple ranging methods and visual image processing .
Ultrasound: A comparative image of the metaphor is the bat. The flying mammal, which emits ultrasound through the special structure of the larynx in the mouth, reflects when ultrasound encounters prey or obstacles and bats can use special hearing systems to receive the reflected signal to detect the target The distance to determine the flight route.
Ultrasound is the most simple ranging system, most of the life encountered in the ranging system is the use of this technology, the most common is the car reversing radar. In the drone equipped with directional ultrasonic transmitter and receiver, and then access the flight control system can be.
However, the application of ultrasonic obstacle avoidance system in UAV also has obvious interference problems. Although the ultrasonic obstacle avoidance system will not be exposed to light, dust and smoke, it will be affected by sound waves in some scenarios. Secondly, if the surface of the object is not capable of reflecting the ultrasonic wave, the effective distance of obstacle avoidance system will be reduced, and the potential safety hazard will be significantly increased. Generally speaking, the effective distance of the ultrasonic wave is 5 meters. The corresponding reflecting object is made of concrete floor. If the material is not a smooth solid, such as a carpet, then the reflection and reception of the ultrasonic wave will be problematic.
For UAVs, this ultrasonic sensor ultrasound system should be placed in multiple directions, such as around the four directions around, you can hover and fly to keep the surrounding to monitor; and placed under the fuselage and above , You can avoid hitting obstacles or the ground too quickly when taking off, descending and landing.
TOF: Ordinarily speaking, it is to replace the front of the ultrasound into light. There are two ways to detect light time, and the other is the phase of light. But in general, the light is shot out, and then detect the light reflected back, and then determine whether there are obstacles around the UAV, geometric distance and so on.
TOF obstacle avoidance system uses optical phase detection. In the same way as the ultrasonic waves, the light waves are also disturbed. At present, the light pollution between buildings in the urban environment poses a problem for the TOF obstacle avoidance system. The light emitted by the system must avoid the main energy band of the sunlight and prevent the sunlight Direct radiation, reflections and other obstacles to the obstacle avoidance system.
As for the effective distance of the TOF system, Currently, TOF measures distances up to 10 meters indoors, and about 5 meters when outdoor glare interferes.
The TOF system keeps spinning fast while hovering, rotating 2-5 turns per second. This is because during the rotation, the system can quickly scan the 360 ° range around the effective radius so as to find the obstacle at a faster speed and then issue instructions to the flight control system to adjust the position so as to prevent the surrounding People or property damage; while in flight, TOF system will stop rotating, only the light to the direction of the forward. When fixed in the direction of the outdoor effective distance can be increased to 8-10 meters. For a typical drones, the wonderful flight distance is about 10 meters, 1 second after the detection of the reaction time of the obstacle, the drone can use a larger acceleration to stop moving forward, which is enough.
For the ultrasonic sensors apply, OSENON’s 40C16T/R-1 is for the detection. Obstacle recognition can be performed in front, back, left, right, and bottom in a total of five directions, and the recognition mechanism is divided into two parts, the ratio of which is ultrasound and machine vision. That is to say, in addition to the conventional ultrasound module, cameras are specially placed in five directions for acquiring visual images and then transmitted directly to an onboard Intel Atom Bay Trail processor for computation. In addition, Parrot, which entered the consumer-grade UAV market earlier in the market, is also working with Nvidia on obstacle avoidance and also uses a hybrid obstacle avoidance system that includes machine vision.