Understanding The Role of Drivers, Switches, and Laser Diodes for Effective LiDAR Performance

Light detection and ranging (LiDAR) systems have become the preferred method for enabling an automobile, an automated guided vehicle (AGV), or even a robotic vacuum to “see” its surroundings. Drones and higher-flying aircraft also use LiDAR to navigate and map terrain at greater distances.

Though LiDAR has been well studied, designers must exercise great care when selecting key components such as the gate driver, the gate-switch FET, and the laser diode necessary to create the optical pulses.

This article provides an overview of LiDAR. It then presents examples of the critical electro-optical components and shows how they work together to generate the necessary pulses.

How LiDAR works

LiDAR operates by sending out a continuous stream of short, moderate-power optical pulses and then capturing their reflections. It measures time-of-flight (ToF) to create a point cloud of the surroundings that presents a three-dimensional (3D) perspective (Figure 1). Many systems use multiple laser diodes in a matrix for broader area coverage.

Figure 1: The LiDAR approach creates a point cloud that provides a 3D rendering of the surroundings. (Image source: Blickfeld GmbH)

The application determines the performance of a LiDAR system. A system used for a slow-moving, area-constrained robotic vacuum or an AGV has much looser range and angular resolution requirements than one used in a car, which must contend with faster speeds and respond to vehicles, cyclists, or pedestrians. The numbers often cited as top-level performance objectives for automotive applications are an effective range of 100 m to 200 m and an angular resolution of 0.1°.

A two-axis electromechanical galvanometer scans the laser flashes across the image area to achieve a precise point cloud. Since the LiDAR system measures ToF for each emitted pulse and its associated return, it can create a 3D image with the depth perspective necessary for vehicles to navigate their surroundings accurately.

The electro-optical path at LiDAR’s core

A complete LiDAR system, such as the one used in an AGV, requires a diverse set of interconnected optical, analog, processor, and mechanical blocks. At the system’s core is the electro-optical path, which comprises a laser-based optical source and a co-located optical receiver (Figure 2).

Figure 2: The electro-optic signal path and associated components are the heart of a LiDAR system (right side, middle row). (Image source: ROHM)

The signal path of the source that creates the stream of optical pulses is controlled by a dedicated microcontroller unit (MCU), which determines the desired optical pulse repetition rate and width. The source path has three key functional elements:

• The gate driver provides high-speed pulses with fast rise and fall times to turn the gate switch on and off.
• The gate-switch FET turns crisply on and off to control the laser diode's current flow.
• The laser diode creates independent, non-overlapping optical pulses at the required wavelength.

Selecting and integrating these components requires an understanding of electrical issues, as well as optical characteristics such as field of view, laser diode power and wavelength angular sensitivity, and optical signal-to-noise ratio (SNR). Advanced software algorithms can overcome some limitations in the electro-optical signal paths and challenges in the sensed setting. However, it is prudent engineering to choose components optimized for LiDAR rather than assume these algorithms can compensate for shortcomings.

A look at a representative component for each of these functions illustrates how LiDAR-optimized devices address the many challenges:

The gate driver

The ROHM Semiconductor BD2311NVX-LBE2 (Figure 3) is a single-channel, ultra-fast GaN gate driver well-suited for industrial applications such as AGVs. It provides the necessary combination of drive current and voltage. It comes in a 6-pin package measuring just 2.0 mm × 2.0 mm × 0.6 mm, and can source up to 5.4 A of output current with a supply voltage range of 4.5 V to 5.5 V.

Figure 3: The BD2311NVX-LBE2 single-channel gate driver provides the necessary combination of drive current and voltage to precisely control a LiDAR gate switch. (Image source: ROHM)

The BD2311NVX-LBE2 can drive GaN high electron-mobility transistors (HEMTs) and other switching devices with narrow output pulses, thus contributing to LiDAR’s long range and high accuracy. These pulse-related parameters include a minimum input pulse width of 1.25 nanoseconds (ns), a typical rise time of 0.65 ns, and a typical fall time of 0.70 ns, all with a 220 picofarad (pF) load. The turn-on and turn-off delay times are 3.4 ns and 3.0 ns, respectively.

The gate-switch FET

The output of the gate driver connects to the control input of the current-control switch device. This device must rapidly switch between on and off states as directed by the gate driver and handle relatively large current values, typically 50 A to 100 A.

The required level of performance is available using devices such as EPC’s EPC2252, an automotive-qualified (AEC-Q101) N-channel, enhancement-mode GaN power transistor. It features exceptionally high electron mobility and a low temperature coefficient for a very low on-resistance (R_DS(ON)), while its lateral device structure and majority carrier diode provide an exceptionally low total gate charge (Q_G) and zero source-drain recovery charge (Q_RR). The result is a device that can handle tasks where very high switching frequency and low on-time are beneficial, and where on-state losses dominate.

The EPC2252’s 80 V drain-source voltage (V_DS), 11 milliohms (mΩ) (maximum) R_DS(ON), and continuous drain current (I_D) of 8.2 A only tell part of the story. It is easy to use, requires an on-state gate drive of only 5 V, 0 V for the off state, and does not need a negative voltage. This simplifies both driver and supply rail considerations.

Due to its design and die arrangement, the gate switch can handle an I_D of 80 A at 125°C (T_PULSE of 10 microseconds (µs)) and is packaged as a passivated die measuring 1.5 mm × 1.5 mm with nine contact solder bumps (Figure 4). Reduced package-and-die parasitics, such as an input capacitance (C_ISS) of 440 pF (typical), support high-speed pulse performance with fast transitions.

Figure 4: The EPC2252 GaN power transistor provides the needed current switching for high-current laser diodes in a package measuring 1.5 × 1.5 mm. (Image source: EPC)

The laser diode

This is the final component in the optical path and functions as an electro-optical transducer. Unlike cameras, which are passive devices, laser diodes are active sources and emit optical radiation, which is deemed harmful to human eyes under some conditions. The maximum allowed intensity is defined by standards such as EN 60825-1:2014, “Safety of Laser Products.”

The safety rating of a LiDAR system depends on its power, divergence angle, pulse duration, exposure direction, and wavelength. Most systems use a 905 nanometer (nm) or 1550 nm wavelength, each offering acceptable efficiency and wavelength compatibility between the laser and a suitable photodiode. Generally, a 1550 nm laser can safely emit more power than a 905 nm laser before it is deemed unsafe. However, 905 nm lasers are popular because they are more cost-effective.

For a 905 nm wavelength, the ROHM RLD90QZW3-00A is a pulsed laser diode optimized for LiDAR applications. It supports 75 W output at a forward current (I_F) of 23 A and provides superior performance across three parameters: beam width (divergence), beam wavelength narrowness, and beam stability.

Beam divergence defines the spread of the beam due to diffraction. The RLD90QZW3-00A specifies typical values of 25° in the perpendicular plane (θ_⊥) and 12° in its parallel plane (θ_//) (Figure 5). Its laser output temperature stability is 0.15 nm per degree Celsius (nm/°C).

Figure 5: The RLD90QZW3-00A pulsed laser diode has typical beam divergence values of 25° in the perpendicular plane (left) and 12° in its parallel plane (right). (Image source: ROHM)

The narrow light-emission width and stability of this laser diode's output wavelength are also critical to enhanced system performance, as they allow the use of narrow-wavelength optical bandpass filters. ROHM states that this diode's 225 micrometer (μm) range is 22% smaller than available competitive devices, thus supporting higher resolution and a wider sensing range with high beam sharpness, narrow emissivity, and high optical density.

These two factors improve the optical SNR, enabling accurate sensing and assessment of objects at an extended distance. A comparative point-cloud image shows the positive impact of these tight and stable specifications on resolution (Figure 6).

Figure 6: The stability and consistency of the RLD90QZW3-00A pulsed laser diode output yields an improved SNR and point cloud resolution. (Image source: ROHM)

Conclusion

LiDAR is widely used to capture 3D perspectives of surroundings and map terrains. At the core of the LiDAR system are the electronic and electro-optical components that integrate the complex capabilities needed for a viable system. For the optical-source functions, the gate driver, gate-switch FET, and laser diode must be compatible with respect to voltage, current, speed, and stability to ensure optimal performance.

Related LiDAR content

• Ensure LiDAR Automotive Distance Sensor Precision With the Right TIA
• An Understanding of How LiDAR Works Shows the Importance of Careful TIA and Comparator Selection
• Simplifying Time-of-Flight Distance Measurements
• Get Started Quickly With 3D Time-of-Flight Applications
• Quick Guide to GaN FETs for LiDAR in Autonomous Vehicles
• Integrated Time-to-Digital Converters Simplify Time-of-Flight Range-Finding Designs