Recently, no driving vehicles have caused a wide range of hot discussions. The industry is working hard to develop this technology as a method of improving security and savings. This is no longer a unreachable idea, BMW, Mercedes and Tesla and other companies have released or will release vehicle systems that can be driven automatically.
Some financial companies are vigorously investing in self-guided vehicle technology. Many of us are familiar with Google Drone Project and Its Test. We are also very likely to be familiar with Amazon's plan for the use of self-guided drone delivery. However, you may not know that food companies such as Domino's Pizza are actively developing electric vehicles that can be delivered in 30 minutes, as well as drone-based pizza delivery systems. In addition, large shopping places such as Wal-Mart are also invested in automatic delivery technology, and plans to provide fast and reliable delivery services with Amazon, providing fast and reliable delivery services in unacceptable.
This article will explore sensor techniques that can be used to create autonomous driving vehicles, and how to archize real-time and mixed signal embedded elements to provide relatively safe and reliable automatic driving systems. All parts and technologies mentioned herein can be found in the Digi-Key website.
GPS
All driver vehicles are largely dependent on sensors. The main sensor technology required by any guiding system is location recognition. Of course, GPS and geographic positioning techniques have embedded in most smartphones. Although these systems are quite reliable, they are still unreliable enough to meet individual unmanned vehicles or fleets of the same road and streets with traditions.
Think about your personal experience. How many GPS positioning is not accurate or good efficiency? If there is no human mind, the weather conditions, electronic noise sources, and maps are inconsistent, and other factors can lead to catastrophic consequences.
Another factor is the resolution. If you only need to be accurate to rice, then today's GPS may already have good enough, but if you need to accurately go to centimeter resolution? In the range of rice meter, the vehicle will collide with the head. Similarly, the meter-level precision incorrectly generated on the coastal road may lead to the cliff of the vehicle.
In addition to the location, the vehicle anti-collision is another key technique for automatic driving a car. Although automatic driving vehicles can be programmed to improve driving vigilance, ensure that they are uncontrusive, music or mobile phone, but they must also share the roads that cannot be safely driving or even abide by traffic rules. This means that the self-directed vehicle is in mind, and must also handle the person's error in combination with the plan and strategy.
Available solutions
For overall position perception, modular GPS solutions are nice choices. Low-cost low-power small GPS modules can support a variety of international agreements, such as China's Beidou Satellite Navigation System, Russia's Glonass, the EU Galileo Satellite Navigation System (GNSS) and the US Global Positioning System (GPS) technology. Antenova M10478-A3 modules include broadband antennas that provide support for all of the above protocol standards, while providing UART-based simple interfaces with up to 115.2 kbits / sec, almost easier to connect any standard embedded microcontroller.
Design engineers should pay attention to the limit factors often do not have the size and power of the vehicle, but the sensitivity and expansion temperature range. The Antenova module has a high sensitivity of -165 dB, its operating temperature ranges between -40 to + 80 ° C. 10 Hz fixed update rate helps ensure that the latest location information is quickly obtained at any time, so that the relatively slow vehicle moves normally on the ground.
The housing module of the rated voltage is 3.3 V, 38 mA itself is mounted on the 13.8 x 9.5 x 1.8 mm SMT mounting plate on the 28-pin, which can be directly reflowed to the main motherboard containing the embedded microcontroller. The development and evaluation board of Antenova M10478-A3-U1 can easily connect USB to quickly test and evaluate the solution while investing less and low risk.
Map memory and data will be licensed or patched with universal map solutions such as Google Maps, and may also require GSM and internet connection. GSM solutions can also provide GPS feature. For example, the Maestro Wireless Solutions M1003GXT48500 3G GSM modem integrates the GSM function and GPS support to enable geopositioning and GPS functionality (Figure 1). This provides an alternate solution that utilizes a honeycomb signal to transmit tower and satellite to get a correction.
Collision
Detecting objects to avoid collisions, is a key part of automatic navigation vehicle safety requirements. Although a large number of CCD image sensors and cameras can be obtained from various channels, if engineers need to develop sensor algorithms quickly and efficiently, the video sensor and control processor should have a flexible platform.
The ideal solution is to detect edges, optimize images, and perform computing in fast hardware using FPGA techniques that can directly support video streams, thereby determining the speed, direction, and approximate level close to the object and threatened evaluation.
The Lattice Semiconductor provides an ideal solution that comes with its LFE3-70EA-HDR60-DKN development system for 1080p, 60 FPS video camera. This platform contains a reference design and IP that can be used with the company's LCMXO2-4000He-DSIB-EVN image interface board and its LF-9MT024NV-EVN Nanovesta camera.
This technique combines two image sensors into a video data stream (Fig. 2), which can perform depth perception and more accurate speed and location sensing, and can achieve automatic white balance, eliminate 2D noise, and provide the fastest automatic exposure of the industry. , Resolution up to 16 megapixels.
Radar components
Miniature radar devices have been widely used in the automotive industry. These devices utilize RF techniques to determine proximity, range, speed, and relative object size.
Because the automotive industry has been widely used in its anti-collision fields, automatic parking systems, radar chips, components, development systems, and sub-assemblies can be used for unmanned vehicles.
Please note that the design of multi-channel radar systems has been completed and used, mainly because vehicles have multiple faces need to be monitored. A single-channel radar system such as a lifting door detection system can also be used, but a set of radars must be installed on each side, ax, and autonomous driving vehicles. For example, the forward radar and lateral radar apply to avoid pedestrians, but autonomous vehicles may also require depth detection to determine that there is a trench in the roadside, tire leakage, etc. Is not suitable for parking here.
The components used in this area may include part of Texas Instruments AFE5401TRGCTQ1. This is a single-chip 4-channel simulated radar front end, equipped with integrated low noise amplifiers, equalizer, programmable gain amplifiers, anti-aliasing, and A / D with 12-bit resolution (Figure 3). It is particularly interesting that the rated voltage of 1.8 volts allows for synchronous sampling at all channels, and the sampling rate is up to 25 million samples per second, and the 12-bit compatible parallel bus can quickly transfer the collected data to the local host. Controller.
Analog Devices provides similar solutions with its AD8285WBCPZ 4 channel front end 12-bit ADC. This part is supported by the company's AD8285CP-EBZ RX path radar evaluation board that flexibly captures data using an integrated FIFO connected to the SPI.
In summary, automated transportation using automatic driving vehicles may be difficult in implementing initial costs, and may face difficulties in various potential disorders, such as norms and responsibilities. However, the necessary technical solutions (especially in the form of sensor-based modules and systems) now or is now available for design engineers. Today, advanced auxiliary driving systems (ADAs, such as adaptive cruise control and automatic emergency brakes) using GPS, radar and video, have shown that autonomous driving vehicles are just late and later.
Our other product:
