Understand Drone Design Trade-Offs Before Piling on the Sensors

By Steve Taranovich

Contributed By Digi-Key's North American Editors

Drones are increasingly finding use in many applications, including as part of a first responder’s tool kit at the scene of an emergency or disaster. For example, during the fire at the Notre-Dame Cathedral in Paris they were used to initially report the size, heat, and extent of the active fire. They were also outfitted with thermal imaging capability to search for people still inside. Later, they were used to assess the damage. Clearly this kind of application presents challenges in trying to see through complex conditions such as smoke and flames with adequate resolution.

As enticing as it may be to add more sensors to a drone to address these challenges, designers need to remain aware that drones are battery powered, and in many cases, cost sensitive. As a result, designers need to perform a delicate balancing act between functionality, cost, size, weight, and power consumption (SWaP). Finding this balance is the primary objective when considering the addition of sensors and imaging equipment to a drone design.

This article discusses the architectural trade-offs designers need to consider when adding sensors to a drone. In doing so, particular attention is paid to the power supply, which will likely have magnetics that can add excess weight and take up precious space. In addition, suitable power supply and sensor solutions are introduced from vendors that include Texas Instruments, Efficient Power Conversion, Analog Devices, Bosch Sensortec, STMicroelectronics, and SparkFun Electronics.

Drone architectural design considerations

The power supply: Once the designer knows the key areas on which to focus for optimum drone performance, they can then look at ways to minimize its physical size and weight, beginning with creating the most efficient power supply possible. This will lead to the minimization of the overall power supply size and weight, and so to a smaller, lighter drone.

Being battery operated, a drone with greater power supply efficiency can operate with a smaller battery size and weight. A typical choice for a drone battery would be a rechargeable lithium battery—Li-Ion or Li-Po type—especially if the designer plans to recharge the battery when landing or hovering over a wireless charger, or just upon landing with an external charger. Designers can also use a standard non-rechargeable battery as the power source and replace it once it is discharged.

When choosing a DC/DC converter, designers will need to use a wide input device due to the high voltage pulse of back EMF (BEMF) from the rotor motors. Under motor deceleration, this BEMF will appear at the DC/DC converter’s input as it comes after the separate DC/DC conversion powering the rotor motors.

The Texas Instruments LM5161 DC/DC power converter IC is a good choice for a drone power supply because when programmed for discontinuous conduction mode (DCM) operation, it provides a tightly regulated buck output without any additional external feedback ripple injection circuit. It also has integrated high-side and low-side MOSFETs which save on board space. For added reliability, the LM5161 has peak and valley current limit circuits which protect against overload conditions. As an added precautionary feature, an undervoltage lockout (UVLO) circuit provides independently adjustable input undervoltage threshold and hysteresis.

There most likely will be many sensors aboard a drone, along with an associated sensor fusion IC, the main processor, and propeller motors. These require a good battery control system.

Designers may opt for gallium nitride (GaN) power transistors in the power supply architecture they choose that normally uses a power transistor. GaN will help with optimum performance efficiency with minimum size/footprint.

Wireless power - Re-charging while hovering [theoretical discussion]:1, 2, 3 This is desirable because when a drone lands and powers down to re-charge and takes off again, the start-up and liftoff of the rotor motors takes a great deal of power from the battery. Efficient Power Conversion is one of many companies researching wireless charging while hovering. An option for the power supply could be a wireless charging architecture based on a GaN FET, such as Efficient Power Conversion’s EPC2019.

GaN-based FETs allow switching at 13.56 megahertz (MHz)—a switching frequency difficult to reach with ordinary silicon FETs. This high switching frequency will also minimize the size and weight of power supply magnetics. In addition, GaN transistors are five to ten times smaller than silicon devices yet can handle the same power levels. With this type of power supply, drones do not have to land; they can instead hover over a wireless charging base.

Designers will find that there are a great many evaluation/development boards to speed time to market with wireless power. In the case of the EPC2019 GaN FET, Efficient Power Conversion supports it with the EPC9513 wireless power receiver development board, to be used inside the drone. This development board is important to designers because it is based on the AirFuel standard, which ensures a certified wireless design that is interoperable with other wireless charging products, globally. Designers can request the Gerber files from the supplier for the demo board to recreate the board’s optimized layout.

Solar power: Another power option is to use solar energy to charge a drone battery. For this purpose, the PT15-75 solar cell from PowerFilm Inc. is a good option.

The PT15-75 can be used in conjunction with an Analog Devices LT3652 battery charger IC to implement a clever, compact battery charger design (Figure 1). Remember, there really is no situation in which open-circuit voltage (Voc) is output when the panel is attached to a load and providing current.

Diagram of Analog Devices LT3652 battery charger ICFigure 1: Designers can create reliable, efficient drone power with the addition of this 2 A solar-powered battery charger where thermistor RNTC has been added to compensate for a solar cell (like the PT15-75) temperature coefficient at maximum power levels. (Image source: Analog Devices)

The LT3652 input regulation loop also has the capability to find the maximum power operating point of the solar panel, which optimizes the efficiency of conversion from the sun’s power to supply maximum output power to the battery.

Sensors: Sensors will both increase the controllability of drones, as well as their usefulness. With regard to controlling the drone, a sensor can enable an auto level mode, a constant altitude mode, or an orbit mode for circling around a specific object or point of interest. All of these added features rely on higher performance inertial measurement units (IMUs) and barometric pressure sensors to achieve an optimal user experience, as well as improved reliability for special purpose or commercial drones.

Designers may need to increase drone performance, which may require a gyroscope with extremely low output signal drift to ensure drone orientation, position, and balance, especially under changing temperature conditions. This can be achieved using a Bosch Sensortec BMI160 accelerometer and gyroscope combination that comes as a small, low-power IMU with nine-axis sensor data fusion. It measures 2.5 x 3.0 millimeters (mm) with a height of 0.83 mm and consumes only 925 microamps (µA), even when the gyroscope and accelerometer are in full operating mode. It operates from a 1.71 volt to 3.6 volt supply.

To complement the BMI160, a digital barometric pressure sensor with temperature sensor will help to measure vertical velocity, enhance GPS navigation, and determine a drone’s altitude. It is recommended that barometers be occasionally calibrated at sea level pressures to stay accurate. Bosch Sensortec’s BMP388 barometric pressure and temperature sensor is a good example of an IC that designers can integrate into their architecture. With a small footprint of 2 x 2 mm2 at 0.88 mm high, and a low power consumption of just 3.4 µA at 1 hertz (Hz), this sensor module is well suited for battery operation. The device has a typical relative accuracy of +/-8 Pa with a typical absolute accuracy of +/-50 Pa that will improve drone hovering and obstacle avoidance capabilities.

To detect motion along multiple axes, the STMicroelectronics’ ISM330DLCTR iNEMO IMU system-in-package (SiP) module combines an accelerometer and gyroscope along with a magnetometer in a monolithic six-axis IC. This kind of configuration enables a drone to maintain horizontal, vertical, and rotational stability while hovering. For applications like professional-grade drone photography and 3D imagery, six-axis gyro stabilization is necessary and is provided by the ISM330DLCTR.

The gyroscope measures and maintains drone orientation. When integrating three accelerometers, each of which are oriented along a different axis, the degree of motion of a drone along any axis can be determined. This will better enable the collection of information regarding the drone’s roll, pitch, and yaw, and then feed this information back to the drone’s proportional-integral-derivative (PID) controller.

The magnetometer will measure the strength and direction of the Earth’s magnetic north field in order to correct its trajectory. Be sure the magnetometer is calibrated frequently; power lines, motors, and any other strong fields emitted from electrical devices can affect it.

Drone movement caused by external forces, like a strong gust of wind, will be detected by the accelerometer and relayed to the PID controller, which in turn adjusts the motors to compensate.

Rangefinders: Landing, hovering, and distance from an object

Drones need to have good sensors to land safely, hover when wirelessly charging, and sense objects to avoid collisions when in motion. This ranging can be performed using sound or light.

Ultrasonic rangefinder sensing: Drone landing, hovering, and ground tracking capabilities can be provided using ultrasonic sensors. When a drone is in the process of landing, it needs to detect the distance from the bottom of the drone to the area in which it is landing. Although GPS and a barometer are part of this control function, accurate distance sensing is the key to a safe landing.

Ultrasonic sensors can also assist in safe hovering and ground tracking, which may need the drone to fly at a fixed height. One such distance ranging sensor for landing assistance, hovering, and ceiling detection is MaxBotix’s MB1010-000 ultrasonic time-of flight (ToF) ranging sensor board.

Understanding ToF

All of these cases need to use the ToF method, which is the time taken for an emitted ultrasonic wave to reach a target, plus the time for the reflected signal to travel back to the drone’s sensor (Figures 2 and 3).

Diagram of ToF during a drone landing, hovering, or wireless chargingFigure 2: Designers will need to understand the concepts of ToF during a drone landing, hovering, or wireless charging. (Image source: Texas Instruments)

Diagram of three phases of ultrasonic ToFFigure 3: The three phases of ultrasonic ToF. Initial transmitted sound (1), silence (2), and received echo (3) for accurate range finding in their drone designs. An understanding of this graphic, coupled with the evaluation board and sensors discussed in this article, can help designers meet the goals of flight stability, collision avoidance, and optimum wireless charging when implementing the hardware suggestions in this section. (Image source: Texas Instruments)

To calculate the distance from the drone to any object, use the equation:

Equation 1

Texas Instruments offers the PGA460PSM-EVM ultrasonic proximity sensing evaluation module that will shorten design time.

LiDAR range sensing: Another means of distance sensing is with the use of light detection and ranging (LiDAR) with pulsed lasers. The information gained from ToF LiDAR systems may be used to create a three-dimensional image. LiDAR technology allows for high accuracy and resolution, and a large coverage area.

Designers can select an optical laser distance ranging sensor such as the SparkFun Electronics SEN-14032, a laser-based optical ranging sensor with a range of 40 m. An external microcontroller will be needed to interface with the sensor via I2C.

There are two primary kinds of architectures used for this kind of LiDAR: a solid-state LiDAR and a motorized 360˚ field of view rotating LiDAR. These both use the same principle, with a laser sending out a beam of light. In the solid-state case a mirror is used to scan, while the scanning rotating disk architecture uses a spinning disk, driven by a motor.

A third type of LiDAR known as flash LiDAR, flashes many short pulses at the same time, uses a camera chip to receive the pulse reflections, and subsequently measures the ToF. Flash LiDAR has very high resolution but is limited to about 30 meters (m).

Sensing the environment

Thermal imaging camera: A thermal imaging camera on a drone will detect heat signatures/temperature from objects and materials and display them as still images or videos. The Notre-Dame fire in Paris was observed and tracked using thermal image cameras. These cameras can detect small differences in heat, sometimes as small as 0.01˚C.

Another important area for drone thermal imaging use is in disaster recovery, such as after an earthquake or severe hurricane, which can leave behind damaged or collapsed structures with people trapped inside (Figures 4 and 5).

Image of a drone’s-eye view of a collapsed buildingFigure 4: A drone’s-eye view of a collapsed building is an important first step a drone would take with a conventional camera. Then, with the use of a thermal imaging camera, it could sense the body heat of those trapped in the rubble. (Image source: IEEE4)

Image of a trapped person taken using a DJI drone during a fire fighter drillFigure 5: Designers now have the tools to locate and save lives in disaster situations. This image of a trapped person was taken using a DJI drone during a fire fighter drill. (Image source: Industrial Equipment News/Menlo Fire UAS/Drone program, via AP)

A good way for designers to start using thermal imaging in a drone is to use something like the 500-0771-01, a micro thermal camera from FLIR Lepton. The camera has a spectral range of 800 nanometers (nm) to 1400 nm, a scene dynamic range of 0˚ to 120˚C, and a nominal power consumption of 150 milliwatts (mW) (operating), 650 mW (during shutter event), and 5 mW (standby).

Humidity, pressure, and temperature sensing: To help determine atmospheric conditions, designers can use Bosch Sensortec’s BME280, a digital humidity, pressure, and temperature sensor with an SPI interface. It’s highly integrated, measuring 2.5 mm x 2.5 mm x 0.93 mm, and consumes as little as 0.1 µA in sleep mode, or up to 3.6 µA when sensing all three parameters.

Accelerate time to market with multi-sensor development kits

The DA14585IOTMSENSOR is a multi-sensor development kit from Dialog Semiconductor that uses environmental sensors from Bosch Sensortec and motion sensors from TDK Invensense. This kit is important for designers because it is a good platform upon which to experiment with and develop drone sensing sensor fusion capabilities and accelerate time to market.

It has a BME680 low-power gas, humidity, pressure, and temperature sensor, as well as an accelerometer, gyroscope, and magnetometer. The DA14585IOTMSENSOR’s sensor fusion capabilities lets designers see how this feature can be used to both get better overall sensing performance, while also extending drone battery life.


Drones present an unusually difficult design challenge of requiring high functionality and long flight times. As with any design, the main tasks the device will be required to perform must be known in order to develop a plan to ensure an optimal architecture that meets the project requirements.


  1. Drones…Up, Up, and Away
  2. Light-Weight Wireless Power Transfer for Mid-Air Charging of Drones Samer Aldhaher, Paul D. Mitcheson, Juan M. Arteaga, George Kkelis, David C. Yates, IEEE 2017
  3. Nonlinear Parity-Time-Symmetric Model for Constant Efficiency Wireless Power Transfer: Application to a Drone-in-Flight Wireless Charging Platform Jiali Zhou, Bo Zhang, Wenxun Xiao, Dongyuan Qiu, Yanfeng Chen, IEEE 2018
  4. DronAID: A Smart Human Detection Drone for Rescue Rameesha Tariq, Maham Rahim, Nimra Aslam, Narmeen Bawany, Ummay Faseeha, IEEE 2018

Disclaimer: The opinions, beliefs, and viewpoints expressed by the various authors and/or forum participants on this website do not necessarily reflect the opinions, beliefs, and viewpoints of Digi-Key Electronics or official policies of Digi-Key Electronics.

About this author

Steve Taranovich

Steve Taranovich is a freelance technical writer with 47 years of experience in the electronics industry. He received an MSEE from Polytechnic University, Brooklyn, New York, and his BEEE from New York University, Bronx, New York. He was also chairman of the Educational Activities Committee for IEEE Long Island. Presently an Eta Kappa Nu Member and an IEEE Life Senior Member. His expertise is in analog, RF and power management with a diverse embedded processing education as it relates to analog design from his years at Burr-Brown and Texas Instruments.

About this publisher

Digi-Key's North American Editors