Lidar System Components for Autonomous Vehicles

Lidar systems will play a vital role in the entire suite of sensors that keep self-driving cars aware of changes in the surrounding environment. Some designers tend to focus on radar for its obvious complexities, and significant research has gone into 5D imaging using radar systems. However, lidar and other vision systems can also provide a full view of the surrounding environment and will be used alongside radar in newer vehicles.

Lidar vs. Radar: Which is Best for New Vehicles?

To say that one of these is better than the other misses the point of having multiple sensors in a vehicle. Different sensors will be ideal for different applications. Automotive radar is useful for target detection and velocity measurements at short (24 GHz) and long (76 GHz) ranges. Meanwhile, lidar systems use infrared laser pulses (currently 905 nm, but may later move to 1500 nm) to create a map of the surrounding environment. Continuous wave lasers are used in coherent lidar for velocity measurements.

Radar modules designed for automotive applications can already be added to a vehicle with a lower price point than lidar systems. Many companies are exploring the use of both technologies in concert to identify targets, create a map of the environment with tagged targets, and construct images of these nearby targets. This data can then be used with computer vision algorithms to distinguish different objects.

The following metrics are used to evaluate the performance of a lidar system:

• Spatial resolution (lateral and longitudinal)
• Temporal resolution
• Detection range
• Scanning rate

This makes the laser source (both the laser itself and the associated electronics) the most critical portion determining performance. First, a laser with lower beam divergence will have higher lateral resolution. The output pulse energy, divergence, and wavelength determine the available detection range. Timing jitter in the laser driver will determine the longitudinal resolution. Finally, the scanning rate determines the temporal resolution, which is important for accurate velocity measurements. A greater discussion on the lasers involved in these systems deserves its own article. Here, we’ll look at some example components for driving your lidar mapping system and receiving reflected laser pulses.

A look at the ecosystem of sensors in autonomous vehicles

Requirement for Lidar System Components

The components used to support a lidar system, at a rather high level, should meet the following basic requirements:

• Real time detection. The electronics that run your system should process data in near real-time, similar to radar used in ADAS systems.
• High pulse repetition rate. This is a function of your laser and your driver. The scanning rate needs to be ~MHz range in order to provide 360 degree scanning with high angular resolution.
• Low power consumption. Your laser driver should be able to provide fast switching with the desired power output with the lowest possible power consumption.
• High detector sensitivity. Higher sensitivity at the desired wavelength allows your system to use a lower intensity pulse. This helps ensure your system can comply with laser safety standards while providing longer usable range (~300 m range is desired).
• Channel number. Newer lidar systems use 64 source-detector channels in an avalanche photodiode array, PIN diode array, or multi-pixel photon counter. Your system will need to interface with these channels to collect reflected pulses.

Let’s take a look at some of the fundamental components you’ll need to build a lidar system:

Texas Instruments LM1020

The LM1020 is a GaN transistor that provides ~1 ns laser pulses with up to 60 MHz repetition rate and 2.5 to 4.5 ns propagation delay. The use of a GaN high electron mobility transistor provides much better performance than a silicon FET in lidar systems. The drain-source resistance of this GaN FET is approximately half that of an equivalent silicon device, meaning conduction losses are approximately 50% lower.

Simplified block diagram for driving a laser diode in a lidar system. From the LM1020 datasheet.

Analog Devices LTC6561

The LTC6561 low noise transimpedance amplifier is ideal for multiplexing 4 channels from an avalanche photodiode array. Note that photodiode arrays normally operate with 64 channels; multiple LTC6561 modules can be combined to interface with more channels. This IC is packaged in a 4 mm x 4 mm QFN package with an exposed pad for thermal management and low inductance. This transimpedance amplifier has low noise and low power consumption:

The amplifier features 74kΩ transimpedance gain and 30µA linear input current range. Using an APD input circuit with a total capacitance of 2pF, the input current noise density is 4.5pA/√Hz at 200MHz. With lower capacitance, noise and bandwidth improve further. Only a 5V single supply is needed and the device consumes only 200mW. [From the LTC6561 datasheet]

Typical application with 4 APD channels. From the LTC6561 datasheet.

Texas Instruments TDC7201ZAXT

As a lidar system uses time-of-flight measurements to create a 3D map of the surrounding environment, the TDC7201ZAXT time-to-digital converter can be used for 4 cm to several km distance measurements without requiring implementation in an external processor. This provides time-to-digital conversion with picosecond accuracy. This component comes in a 4 mm x 4 mm 25 ball nFBGA package.

Simplified block diagram for time-to-digital conversion in a lidar system. From the TDC7201 datasheet.

The components presented here are just a subset of everything you’ll need for an advanced lidar system. Octopart is here to give you access to a huge range of lidar system components for automotive systems. Try using our Part Selector guide to determine the best option for your next product.

If you found this article interesting, be sure to stay up-to-date with our latest articles by signing up for our newsletter.