LidarBoX: a 3D-printed, open-source altimeter system to improve photogrammetric accuracy for off-the-shelf drones
Abstract
Drones provide a privileged birds’-eye view for collecting high-resolution imagery for morphometric and behavioral sampling of animals. Biologically meaningful measurements extracted from overhead images require an accurate estimate of altitude, but current commercial drones include inaccurate barometer estimates. Recent proposals for coupling altimeter systems to drones have provided customized, open-source solutions, yet assembling such altimeter systems requires advanced technical skills, thereby potentially limiting their use. Here, we built upon recent advances to provide a 3D-printed enclosure for an altimeter system that is inexpensive, self-contained, easy to setup, and transferable across commercial drones. We depart from a published, successful data logger system composed of a GPS and LiDAR sensor and design a more compact and self-powered version (“LidarBoX”) that easily attaches to a variety of commercial drones. We compare flight times with/without LidarBoX attached, test flight maneuverability and performance, and validate the reliability of measurement accuracy. To make LidarBoX accessible, we provide an open-source repository with design code and files and a how-to-assemble guide for non-specialists. We hope this work helps popularize LiDAR altimeter systems on commercial drones to improve the accuracy and reliability of drones as a sampling platform for ecology and wildlife research.
1. Introduction
The recent advancement in drones, or unoccupied aircraft systems (UAS), has greatly enhanced opportunities for scientists across a broad range of disciplines to collect high-resolution aerial imagery. Wildlife researchers in particular have utilized this technology as an alternative to the high costs of crewed aircraft for mapping and assessing habitat quality (Olsoy et al. 2018), estimating animal population size, density, and abundance (Vermeulen et al. 2013; Chabot et al. 2015; Seymour et al. 2017; Bird et al. 2020; McMahon et al. 2022) tracking individual behavior and movement (Torres et al. 2018; Graving et al. 2019; Oleksyn et al. 2021; Ejrnæs and Sprogis 2022), and monitoring the health of free-ranging individuals (Pirotta et al. 2017; Horton et al. 2019; Lemos et al. 2020, 2022; Stewart et al. 2022; Leslie et al. 2023).
Aerial photogrammetry has a long history of providing a non-invasive method for obtaining morphological measurements of free-living large elusive megafauna, particularly cetaceans (Whitehead and Payne 1978; Cubbage and Calambokidis 1987; Best and Rüther 1992; Perryman and Lynn 1993). As such, drones greatly increase accessibility and efficiency in collecting aerial imagery of marine megafauna for morphometric analysis (Johnston 2019). For example, photogrammetry from drone-based imagery has been used to measure the body size of tunas, manta rays, sea turtles, sharks, manatees, pinnipeds, and cetaceans (Jech et al. 2020; Durban et al. 2021; Shero et al. 2021; Piacenza et al. 2022; Ramos et al. 2022; Setyawan et al. 2022; Vivier et al. 2023). Drone-based photogrammetry is particularly useful for estimating the body condition of baleen whales (Christiansen et al. 2018; Bierlich et al. 2021a, 2022; Torres et al. 2022), and has been used to assess morphological differences between blue whale populations (Balaenoptera musculus) (Leslie et al. 2020) and killer whale (Orcinus orca) ecotypes (Durban et al. 2021; Kotik et al. 2022), compare the biomechanics and energetics of foraging across different baleen whales species (Kahane-Rapport et al. 2020; Gough et al. 2022; Segre et al. 2022), detect pregnancy in bottlenose dolphins (Tursiops truncatus) (Cheney et al. 2022) and gray whales (Eschrichtius robustus) (Fernandez Ajo et al. 2023), and even identify stunted growth in killer whales (Groskreutz et al. 2019) and North Atlantic right whales (Eubalaena glacialis) (Stewart et al. 2021, 2022).
Acquiring accurate altitude is a crucial component for obtaining accurate morphometric measurements, as altitude is a key variable in setting the scale of the image (Bierlich et al. 2021b). All drones are equipped with a barometer, which measures altitude based on changes in pressure. In general, barometers usually yield low accuracy in the altitude recorded, particularly for low-cost sensors commonly found on small, off-the-shelf drones (Wei et al. 2016; Bierlich et al. 2021b). Dawson et al. (2017) developed a data logger for off-the-shelf drones containing a small, lightweight Light Detection and Ranging (LiDAR) laser altimeter that yields greater accuracy in the recorded altitude. LiDAR altimeters use light in the form of a pulsed laser and measure reflected pulses to obtain the direct distance to a target. Recent experiments compared the accuracy of measurements when using a barometer versus an attached LiDAR system (Bierlich et al. 2021b; Ramos et al. 2022) and confirmed that (i) large variation in measurement error exists across different drones when using altitude from a barometer, and (ii) using altitude from an attached LiDAR system dramatically reduces measurement error across drone types (professional, off-the-shelf, and customized). Therefore, as drones continue to be widely used in wildlife research, it is crucial for wildlife researchers to integrate LiDAR systems on their drones to obtain accurate morphological measurements. Several studies have adopted the LiDAR system introduced by Dawson et al. (2017), installing customized versions for collecting morphometric data on different species of cetaceans (Christiansen et al. 2018; Bierlich et al. 2021a; Cheney et al. 2022; Ramos et al. 2022). However, assembling and installing a LiDAR system on off-the-shelf drones can be difficult compared to more customizable drones and requires technical and electrical training.
Here, we introduce “LidarBoX”, a 3D-printed enclosure for an inexpensive, self-contained, easy-to-setup LiDAR system that is transferable across off-the-shelf drones. We build off the successful data logger system composed of a GPS receiver and a LiDAR altimeter created by Dawson et al. (2017) and design a more compact version of the system that can be swapped between commercially available drone models. In what follows, we describe the implementation of the inner electrical components, the design of the external 3D-printed enclosure, and provide instructions for safely assembling and using the LidarBoX system. We then validate the functionality of the LidarBoX and install it on two commercial-grade drones, a DJI Phantom 4 Pro and a DJI Inspire 2, to test flight maneuverability, performance, and battery usage (flight time). Finally, we test the reliability of the LidarBoX by comparing it to the Dawson et al. (2017) LiDAR system using calibration flight measurements.
2. Overall implementation and design of LidarBoX
The design of the LidarBoX includes a 3D-printed housing, 2-layer printed circuit board (PCB) design, a laser-cut housing lid, and the full system build. To develop the hardware of the LidarBoX, we arranged the system components (Table 1; Dawson et al. 2017) in a compact manner, considering physical size and specific location requirements needed by the microSD card and LiDAR assemblies. Once satisfied with the electronics layout, we designed a custom 3D-printed enclosure to house the electronics package (Fig. 1A). We incorporated a PCB to reduce assembly time and increase reliability by ensuring proper component placement and eliminating point-to-point wiring (Figs. 1B and 1C). Once the altimeter system enclosure was arranged, we 3D-printed a platform to mount the enclosure on the drone (Figs. 1A and 1C). We attached the platforms to the drone by replacing the original posterior underside screws on the shell of the drone with two longer ones for the Phantom 4 Pro and reusing the screws for the Inspire 2. Finally, a 7.4 V Li-Ion battery 500mAh pack was inserted on top of this platform and secured through Velcro bands (Figs. 1B and 1C). We provide the printing design files for both the system enclosure and the attaching platforms designed for two commercially available drone models: https://doi.org/10.7267/3n204680g.
Fig. 1.
Table 1.
2.1. Details on assembly
The PCB was designed so that all connections are easily soldered by hand and utilize standard 0.10 in. pin spacing for headers. The Pro Micro used for control in this setup (Table 1) utilizes a 24-pin DIP socket to plug into the board, making it easy to replace. Once all components are soldered onto the PCB wiring diagram, a multimeter should be used to check continuity paths on the board. After testing, the Pro Micro can be plugged into a computer, and code can be uploaded via the Arduino IDE and the GitHub repository from Dawson et al. (2017). Depending on previous setups, additional steps may be needed to connect the Pro Micro to the Arduino IDE; these steps are found within the board data sheets and through the Sparkfun setup guide (see Table 1). No modification of the Dawson et al. (2017) code library is needed, and it is recommended to connect directly to the LiDAR unit to enable the median filter (as outlined in Dawson et al. 2017).
2.1.1. Availability of materials
All parts and materials to build the LidarBoX altimeter system have been sourced from online vendors (see Table 1) and can be purchased in single quantities. The enclosure can be built using professional or personal-grade 3D printers using standard polylactic acid material, yet stronger materials such as PET-G or ASA are recommended. Electronic CAD software “Eagle” by Autodesk was used to design the printed circuit board. The only critical component on the PCB is the SD card port. Its placement had to be as close as possible to an outside edge to ensure the ease of the insertion and removal of the SD card. The build files were sent to a local fabrication house, where a professionally printed circuit board was manufactured. Less expensive options, including DIY solutions, are possible.
2.1.2. Ease of build
The LidarBoX parts are common and easily available from online vendors, while the build of the system enclosure requires a 3D printer. Extensive use of assemblies, break-out boards, and off-the-shelf components, along with detailed silk screening of the PCB with component locations, make assembly of the electronics easy, even for beginners. Common hand tools (e.g., hex keys, screwdrivers, and pliers) and modest soldering skills are all that are required. Alternately, users can arrange to have a LidarBoX, or components of the LidarBoX (such as the printed circuit board), built for them by the Oregon State University's Innovation Lab (iLab; https://hmsc.oregonstate.edu/iLab). Contact the corresponding authors for more information.
2.2. Quality control
2.2.1. Safety
Builders of the LidarBoX altimeter system should be aware of potential hazards such as electric shock or burning when assembling electronics or cutting out the plastic cover. Testing a newly developed circuit incurs the risk of electric shocks, so care should be taken: all cables and sensors should be checked for damage, and the workspace should be at no risk of splashing water. To avoid burning accidents during soldering, we advise adhering to the recommendations of the user's manual. To avoid accidents with the 3D print, we recommend PET-G. Standard personal protective equipment (PPE) should be worn during all stages of the system assembly.
Users of the LidarBoX altimeter system should be aware of potential hazards when attaching the enclosure to the drone (i.e., pinched fingers) and injury from the drone propellers during operation. When flying drones, we strongly recommend that the pilot be properly trained and certified and to follow all the necessary safety protocols (Fiori et al. 2017; Raoult et al. 2020). Contact with drone propellers and impact from a drone collision can cause serious injury or even death (Koh et al. 2018; Moskowitz et al. 2018). Drone operations in marine environments pose additional risks, as hand launch and recovery are often the safest methods to offset the pitch and roll of the vessel and ensure the drone is at a safe distance from other personnel (Raoult et al. 2020). As such, we recommend the use of PPE, such as protective Kevlar gloves and safety helmets with a face shield, when hand-launching and recovering the drone to avoid any injuries by the drone propellers. We also recommend having additional personnel with first aid and basic life support training to assist in an emergency.
2.2.2. Validation
As mentioned above, the accuracy of the laser altimeter (LightWare LiDAR SF11/C) presented by Dawson et al. (2017) has been further corroborated by various studies over land and water (i.e., Christiansen et al. 2018; Bierlich et al. 2021b, 2022; Kotik et al. 2022; Ramos et al. 2022; Stewart et al. 2022). To ensure that each component of the system properly functions within our adapted “LidarBoX”, we conducted a series of simple tests. To validate the altimeter system, we first powered it on by connecting the battery pack and confirmed the system was recording data on the microSD card. We checked the precision of the internal GPS clock by comparing it with that of a high-precision external GPS unit (e.g., Bad Elf GPS Pro). Next, we double-checked whether the LiDAR laser sensor was well leveled. To confirm data successfully recorded within our 3D-printed enclosure, we elevated the LidarBoX to five known heights (between 1.164 and 2.909 m) using a ladder and compared the distance (altitude) measured by the LiDAR sensor to the known, measured height at a specific timestamp. We used a Bad Elf GPS Pro to sync the correct time of the altitude recorded by the LidarBoX for each known height on the ladder at the specified timestamp. The LidarBoX had a mean % error of 0.82% (range: −3.8% to 3.45%, n = 12 measurements) when compared to the known height, confirming that the altitude was successfully recorded and estimated with high accuracy.
2.3. Before take-off and post-processing recommendations
The LidarBoX is powered on by plugging into the 7.4 V Li-ion battery 500 mAh pack (Figs. 1A–1C), which activates a solid red light (Fig. 1C). Once the GPS has connected, a yellow light will begin flashing, and data will begin recording. Note that you may need to be outdoors and away from any buildings or tall trees to connect to the GPS. Altitude (or distance) and associated metadata are directly recorded to the microSD card within the LidarBoX.
It is essential to align the collected imagery time with the recorded GPS time of the LiDAR to match the corresponding altitude for a specific image or snapshot, as the timestamps of cameras are susceptible to drift, resulting in inaccurate recorded times (Dawson et al. 2017; Raoult et al. 2020). Thus, recording a timestamp is crucial before each takeoff to ensure that the correct altitude values are matched to the correct time of the snapshot or image used for photogrammetry in post-processing. To sync the recorded LiDAR time with the imagery time, we strongly recommend collecting photos or videos prior to take-off of the time stamp from a high-precision external GPS unit (such as a Bad Elf GPS Pro; www.bad-elf.com) connected and displayed on a phone or tablet screen. This ensures the correct time stamp of an image can be linked to the correct altitude recorded by the LidarBoX by accounting for the offset between the recorded timestamps. The general process of linking the LiDAR values to the images involves (i) calculating the offset per flight using the image of the high-precision external GPS unit's time, (ii) applying the offset to all images, and (iii) merging the image data with the LiDAR data. If images were captured using the drone, this is a straightforward process; however, if snapshots are extracted from video, this process also requires linking the snapshot timestamp from within the video to real-time data and then applying the correction.
Since the LidarBoX placement on the drone is not gimbaled, altitude readings are susceptible to bias due to the tilt of the drone. We correct for this bias in tilt by following the methods described by Dawson et al. (2017), where the tilt degree (t) measured by the inertial measurement unit is used to calculate a corrected altitude (Ac):
where AL is the raw, empirical altitude recorded by the LiDAR and the Cd is the distance of the camera to the laser. While the LiDAR altimeter improves the accuracy of altitude measurements, these recorded distances are still susceptible to null or erroneous values. For example, Bierlich et al. (2021b) recorded that 17.7% of aerial images used in their analysis had null laser altimeter values. Thus, analysts should check for suspiciously low altitudes compared to the normal flight range (i.e., <15 m when most flights are typically between 20 and 40 m) or altitudes greater than 120 m, which are considered null values since they exceed the maximum recording height for the LightWare SF11/C laser altimeter.
(1)
Note that the vision positioning system (VPS) on the drones must be disabled before flight, as the LidarBoX may obstruct these sensors, which can cause the drone to fly in an irregular up-and-down motion, or a “yoyo effect”.
2.3.1. Operating software and peripherals
The LidarBoX system is self-contained, and all necessary data are recorded locally as comma-separated value files (.csv) files in an SD card. Using the LidarBoX for photogrammetry, animal behavior, and wildlife ecology usually requires a posteriori processing of the raw LidarBoX data to sync with other multimedia or text data collected simultaneously, such as audio recordings, photographs for individual identification, and field notes. This can be accomplished by aligning the time-stamped text data exported from the LidarBoX with the time-stamped metadata of other multimedia or field notes, either manually or aided by specific software. One way of doing so is by using different open-source software to extract the raw metadata from media files and then align and sync the media of interest with the GPS position and altitude data recorded by the LidarBoX. The open-source software ExifTool (https://exiftool.org) can be used to extract the metadata of any media file type (e.g., video duration and create date); and the open-source FFmpeg software (https://ffmpeg.org) can be used to clip video and audio files, as well as to extract the embedded subtitle (.SRT files) from videos recorded with DJI drones, to retrieve useful flight metadata (e.g., altitude from the barometer, drone position, and speed). Combined, these tools provide the basics for manually aligning and syncing drone videos (and other media) with the LidarBoX data. This final step can be done by using the open-source R package “MAMMals: Managing Animal MultiMedia: Align, Link, Sync” (Machado and Cantor 2022), which integrates ExifTool and FFmpeg to streamline the processing of multimedia data with time-stamped text data, including that coming from the LidarBoX.
3. Methods
3.1. General testing
We installed the LidarBoX on two DJI drone models—Phantom 4 Pro (P4Pro) and Inspire 2 (I2)—and tested flight performance, flight time, and measurement accuracy (Figs. 1D and 1E). The I2 was fitted with a Zenmuse X5 camera with a Micro Four Thirds sensor (17.3 mm × 13 mm), 3840 × 2160-pixel resolution, and a 25 mm focal length lens (35 mm film equivalent: 50 mm). The P4Pro had a 13.2 mm × 8.8 mm sensor, a 3840 × 2160-pixel resolution, and an 8.8 mm focal length lens (35 mm film equivalent: 24 mm).
3.2. Maneuverability and performance
To test if the LidarBoX and battery remain securely attached to the drones during flight, we performed flights over an open grass area in which the drone was maneuvered, accelerated, and slowed down quickly through all three axes. We accelerated each drone from a stationary, hovering position at 30 m altitude to 10 m/s, covering about 50 m until we brought each to a full stop. We then repeated this process along the z-axis, accelerating from 15 m to 50 m altitude at 10 m/s. Finally, we performed a flight with the same average speed but with repeated rapid changes in yaw, roll, and pitch. We repeated these flights under low (2 m/s) and high wind conditions (7 m/s). The LidarBoX remained securely attached to the drone at all times, demonstrating its resistance to extreme flying. Although we do recommend checking that all screws are fully tightened prior to each flight and carrying additional screws as backup.
3.3. Flight time
We then tested the impact of adding the LidarBoX to the drone on flight time due to the added weight of the enclosure and mounting board (237 g). We compared the flight time (measured here as the time for the drone battery to go from fully charged to 50% charged) under two treatment conditions: (1) with LidarBoX attached and (2) without LidarBoX attached (control treatment). Under the two treatments, each drone was launched, ascended to 15 m, and then hovered in place until the battery reached 50%. The battery percentage was sampled 10 times per second. We used a one-way ANCOVA to compare the flight times of each drone with the LidarBoX attached and unattached.
3.4. Reliability
Since several studies have demonstrated the improved accuracy in measurements when using altitude obtained from the LiDAR altimeter system described by Dawson et al. (2017) (i.e., Bierlich et al. 2021a, 2021b, 2022; Ramos et al. 2022; Stewart et al. 2022; Torres et al. 2022), our aim is to compare the reliability of LiDAR readings made within the LidarBoX relative to the Dawson et al. (2017) system. We compare measurement data for the same calibration object during drone flights using altitude readings recorded by each system. We use a dataset of calibration objects of known size from a tangential research project studying the morphology of Pacific Coast Feeding Group gray whales off the coast of Newport, Oregon, USA, 2016–2022 (Bierlich et al. 2023). In 2021, an I2 drone was used with a LiDAR altimeter attached, following Dawson et al. (2017), whereas in 2022, a LidarBoX was instead attached to the same I2 drone. Following a similar approach to Bierlich et al. (2021b), we compare the accuracy of the calibration measurements when using altitude from a LiDAR system following Dawson et al. (2017) versus a LidarBoX by flying over a known-sized object (1 m wooden board) between 17 and 62 m (Table 2). We used the free and open-source VLC media player (available at www.videolan.org) to extract snapshots of the wooden board in the center of the frame at various altitudes from aerial videos. Each snapshot of the board was imported and measured in pixels using MorphoMetriX open-source photogrammetry software (Torres and Bierlich 2020), with outputs collated using CollatriX (Bird and Bierlich 2020). Length measurements in pixels (Lp) are scaled to standard units (e.g., meters) by
where L is the length measurement in meters, a is the altitude of the drone (m), f is the focal length of the camera (mm), Sw is the sensor width of the camera (mm), and Iw is the image width (px) (Torres and Bierlich 2020). To account for the distance between the camera on the I2 drone and the LidarBoX, we subtracted 3.5 cm. Using the measured length and the true length of the wooden board, we then calculated the % error (% Error) as:
where L is the measured length in meters and Ltrue is the true length of the board. We then tested if the % error between measurements using altitude from the LidarBoX was similar to the LiDAR system from Dawson et al. (2017) using a Student's t test.
(2)
(3)
Table 2.
4. Results
4.1. Flight times and performance
While the LidarBoX did not impact flight performance, both drones had statistically significant reductions in flight times with the LidarBoX attached, P4P: F2, 15963 = 10150, p < 0.001; I2: F2, 15259 = 8622, p < 0.001 (Fig. 2). However, this difference may not be significant in practice depending on flight objectives, as the I2 experienced a 0.47 min difference (−4.19%) (with LidarBoX: mean = 10.75 min, SD = 0.06 min, n = 128; and without LidarBoX: 11.22 min, SD = 0.06, n = 120) and the P4Pro experienced a 2.55 min (−18.33%) difference in flight time at 50% battery (with LidarBoX: mean = 11.36 min, SD = 0.04 min, n = 80; without LidarBoX: 13.91 min, SD = 0.03 min, n = 60) (Fig. 2).
Fig. 2.
4.2. Reliability of LidarBoX
The median % error for measurements using the LiDAR system from Dawson et al. (2017) was −0.18% (n = 51, sd = 2.85%, min = −10.22%, max = 5.91%). The median % error for measurements using the LidarBoX was 0.41% (n = 56, sd = 3.06%, min = 11.59%, max = 4.85%). As such, there was no significant difference between measurements when using the LiDAR system from Dawson et al. (2017) versus the LidarBoX, t(104.94) = 0.83, p = 0.41 (Fig. 3). These findings demonstrate that the LidarBoX is comparable to the LiDAR system from Dawson et al. (2017).
Fig. 3.
5. Discussion
We present LidarBoX, an adapted altimeter system for off-the-shelf drones that is inexpensive, self-contained, and transferable across drone models. By building upon custom systems (Dawson et al. 2017), we provide the template for 3D printing a small and lightweight enclosure, as well as guidance on how to assemble the self-powered data logger system composed of commercially-available parts that can be ordered from major online vendors—a laser altimeter sensor, a GPS unit, and a microSD card reader slot. We show that LidarBoX is lightweight and safe to use, causing only minor, if any, interference to a normal flight profile (e.g., slightly higher battery usage). This system is capable of recording important geographical positions and altitudes accurately, and these data are immediately available for download from a microSD card. Finally, we used calibration data of known-sized objects to demonstrate how measurements using the altitude from the LidarBoX are equivalent to measurements using the altitude from the LiDAR system from Dawson et al. 2017 (Fig. 3).
While LidarBoX can easily be installed and swapped between the two tested commercial drone models to improve the accuracy of altitude recordings, the I2 is much more expensive than a P4Pro. Thus, the lower cost of a P4Pro with the addition of a LidarBoX may serve as an alternative option for research projects on a tight budget. While the P4Pro in general has a longer flight time than the I2, the weight of the LidarBoX had a greater relative reduction in flight time (∼18%) compared to the I2 (∼4%) (Fig. 2). This is likely due to the relatively higher payload on the smaller airframe of the P4Pro compared to the I2, which would result in a greater rate of energy consumption (Figs. 1D and 1E). However, the batteries used for the P4Pro were older compared to the I2 (4 years of intensive use compared to 2 years) and thus may not be able to maintain charge throughout a flight as well as the newer batteries of the I2. Future studies should explore the effect of battery age on flight time to help determine optimal battery usage and lifespan. It is also important to note that the reduction in flight times we present may differ for flights during data collection in the field based on objectives, flight operations, field conditions, weather, behavior of the species, age and care of equipment, and the minimum battery percentage threshold for landing.
In addition to obtaining similar measurements when using altitude from Dawson et al. (2017) and LidarBoX, the % error values of the calibration measurements using LiDAR altitude are similar to other studies (Bierlich et al. 2021b). While we were unable to compare calibration measurements using the LiDAR system from Dawson et al. (2017) and the LidarBoX for the P4Pro, we expect similar results as the I2, where no significant difference between measurements using two altimeter systems was observed. However, as previous studies have pointed out, there is greater overall measurement error associated with DJI Phantom models compared to other custom and commercial drones, including the DJI I2 (Burnett et al. 2018; Bierlich et al. 2021b; Torres et al. 2022).
5.1. Reuse potential and adaptability
The LidarBoX was designed to be self-contained and transferable across different drone models. We tested the LidarBoX on two widely-used off-the-shelf drones (P4Pro and I2), but this design can be reconfigured to work well with other commercially available (i.e., DJI Mavic) or custom-built drones. While the LiDAR altimeter system itself would not change for adaptation to other drone platforms, this may require a redesign of the attaching platform and/or the box enclosure, as well as designing additional supportive structures, such as catching handles (if hand launch and recovery are required), for smaller drones, i.e., DJI Mavic. Reducing the size of the enclosure would also reduce the weight of the system to help maximize flight time.
Future improvements to the box design could also include its location on the drone, as the current placement causes interference with the VPS sensors, requiring it to be disabled during flight to avoid the “yoyo effect”. We recommend the drone pilot gain experience flying with the VPS sensors disabled before conducting field work, which are typically disabled when launching and recovering from boats. Future work could experiment with attaching platforms to other parts of the drone to place the altimeter system enclosure further away from the VPS sensors.
While we present LidarBoX in the context of obtaining morphological measurements of megafauna from drones, these methods can be applied to accurately measure distances in drone imagery to address other research questions in wildlife ecology, e.g., spatial position, inter-individual distances, and fine-scale behavior of animals (Graving et al. 2019), as well as other photogrammetry applications, e.g., surveying and mapping (Olsoy et al. 2018). While the LidarBoX is designed to contain a LiDAR sensor, future designs could explore the use of other sensors, such as thermal and multispectral cameras. The LidarBoX could also be used independently from the drone for obtaining accurate distances of objects as an unattached laser range finder device.
As the use of drones continues to develop and evolve in the wildlife sciences, we hope accessible and adaptable open-source tools, such as LidarBoX, encourage researchers to use and share innovative methods to help improve data collection and yield more accurate measurements.
Acknowledgements
We are grateful for the logistical support from the Innovation Lab (iLab) and the Marine Mammal Institute's Center of Drone Excellence (CODEX), both at the Hatfield Marine Science Center at Oregon State University, to develop the LidarBoX prototype. We thank S.M. Dawson, M.H. Bowman, E. Leunissen, and P. Sirguey for making the computer code and assembly instructions for their data logger freely available; Fabio Daura-Jorge and Alexandre Machado for carrying out preliminary field tests of the LiDAR system with the Phantom4 v2 drone; and the Geospatial Ecology of Marine Megafauna (GEMM) Lab's Gray Whale Response to Ambient Noise Informed by Technology and Ecology (GRANITE) field team for helping with logistics to collect data on Pacific Coast Feeding Group gray whales.
References
Best P.B., Rüther H. 1992. Aerial photogrammetry of southern right whales, Eubalaena australis. J. Zool. 228(4): 595–614.
Bierlich K.C., Hewitt J., Bird C.N., Schick R.S., Friedlaender A., Torres L.G., et al. 2021a. Comparing uncertainty associated with 1-, 2-, and 3D aerial photogrammetry-based body condition measurements of baleen whales. Front. Mar. Sci. 8.
Bierlich K.C., Schick R.S., Hewitt J., Dale J., Goldbogen J.A., Friedlaender A.S., Johnston D.W. 2021b. Bayesian approach for predicting photogrammetric uncertainty in morphometric measurements derived from drones. Mar. Ecol.: Prog. Ser. 673: 193–210.
Bierlich K.C., Hewitt J., Schick R.S., Pallin L., Dale J., Friedlaender A.S., et al. 2022. Seasonal gain in body condition of foraging humpback whales along the Western Antarctic Peninsula. Front. Mar. Sci. 9(1036860): 1–16.
Bierlich K.C., Kane A., Hildebrand L., Bird C.N., Fernandez Ajo A., Stewart J.D., et al. 2023. Downsized: gray whales using an alternative foraging ground have smaller morphology. Biol. Lett. 19(8):.
Bird C., Bierlich K.C. 2020. CollatriX: a GUI to collate MorphoMetriX outputs. J. Open Source Software, 5(51): 2323–2328.
Bird C.N., Dawn A.H., Dale J., Johnston D.W. 2020. A semi-automated method for estimating adélie penguin colony abundance from a fusion of multispectral and thermal imagery collected with unoccupied aircraft systems. Remote. Sens. 12(22): 1–14.
Burnett J.D., Lemos L., Barlow D., Wing M.G., Chandler T., Torres L.G. 2018. Estimating morphometric attributes of baleen whales with photogrammetry from small UASs: a case study with blue and gray whales. Mar. Mammal Sci. 35(1): 108–139.
Chabot D., Craik S.R., Bird D.M. 2015. Population census of a large common tern colony with a small unmanned aircraft. PLoS One, 10(4).
Cheney B.J., Dale J., Thompson P.M., Quick N.J. 2022. Spy in the sky: a method to identify pregnant small cetaceans. Remote Sens. Ecol. Conserv. 8(4): 492–505.
Christiansen F., Vivier F., Charlton C., Ward R., Amerson A., Burnell S., Bejder L. 2018. Maternal body size and condition determine calf growth rates in southern right whales. Mar. Ecol.: Prog. Ser. 592: 267–281.
Cubbage J.C., Calambokidis J. 1987. Size-class segregation of bowhead whales discerned through aerial stereo-photogrammetry. Mar. Mammal Sci. 3(2): 179–185.
Dawson S.M., Bowman M.H., Leunissen E., Sirguey P. 2017. Inexpensive aerial photogrammetry for studies of whales and large marine animals. Front. Mar. Sci. 4(366).
Durban J.W., Fearnbach H., Paredes A., Hickmott L.S., LeRoi D.J. 2021. Size and body condition of sympatric killer whale ecotypes around the Antarctic Peninsula. Mar. Ecol.: Prog. Ser., 677: 209–217.
Ejrnæs D.D., Sprogis K.R. 2022. Ontogenetic changes in energy expenditure and resting behaviour of humpback whale mother–calf pairs examined using unmanned aerial vehicles. Wildl. Res. 49(1): 34–45.
Fernandez Ajo A., Pirotta E., Bierlich K.C., Hildebrand L., Bird C.N., Hunt K.E., et al. 2023. Assessment of a non-invasive approach to pregnancy diagnosis in gray whales through drone-based photogrammetry and faecal hormone analysis. R. Soc. Open Sci. 10(230452).
Fiori L., Doshi A., Martinez E., Orams M.B., Bollard-Breen B. 2017. The use of unmanned aerial systems in marine mammal research. Remote. Sens. 9(6): 513–543.
Gough W.T., Cade D.E., Czapanskiy M.F., Potvin J., Fish F.E., Kahane-Rapport S.R., et al. 2022. Fast and furious: energetic tradeoffs and scaling of high-speed foraging in rorqual whales. Integr. Org. Biol. 4(1).
Graving J.M., Chae D., Naik H., Li L., Koger B., Costelloe B.R., Couzin I.D. 2019. DeepPoseKit, a software toolkit for fast and robust animal pose estimation using deep learning. Elife, 8: p.e47994.
Groskreutz M.J., Durban J.W., Fearnbach H., Barrett-Lennard L.G., Towers J.R., Ford J.K.B. 2019. Decadal changes in adult size of salmon-eating killer whales in the eastern North Pacific. Endanger. Species Res. 40: 183–188.
Horton T.W., Hauser N., Cassel S., Klaus K.F., de Oliveira T.F., Key N. 2019. Doctor drone: non-invasive measurement of humpback whale vital signs using unoccupied aerial system infrared thermography. Front. Mar. Sci. 6(Jul): 1–11.
Jech J.M., Johnson J.J., Lutcavage M., Vanderlaan A.S.M., Rzhanov Y., LeRoi D. 2020. Measurements of juvenile Atlantic Bluefin Tuna (Thunnus thynnus) size using an unmanned aerial system. J. Unmanned Veh. Syst. 1–39.
Johnston D.W. 2019. Unoccupied aircraft systems in marine science and conservation. Ann. Rev. Mar. Sci. 11(1): 439–463.
Kahane-Rapport S.R., Savoca M.S., Cade D.E., Segre P.S., Bierlich K.C., Calambokidis J., et al. 2020. Lunge filter feeding biomechanics constrain rorqual foraging ecology across scale. J. Exp. Biol. 223(20).
Koh C.H., Low K.H., Li L., Zhao Y., Deng C., Tan S.K., et al. 2018. Weight threshold estimation of falling UAVs (unmanned aerial vehicles) based on impact energy. Transp. Res. Part C Emerg. Technol. 93, pp. 228–255.
Kotik C., Durban J.W., Fearnbach H., Barrett-Lennard L.G. 2022. Morphometrics of mammal-eating killer whales from drone photogrammetry, with comparison to sympatric fish-eating killer whales in the eastern North Pacific. Mar. Mammal Sci. 42–58.
Lemos L.S., Burnett J.D., Chandler T.E., Sumich J.L., Torres L.G. 2020. Intra- and inter-annual variation in gray whale body condition on a foraging ground. Ecosphere, 11(4): p.e03094.
Lemos L.S., Olsen A., Smith A., Burnett J.D., Chandler T.E., Larson S., et al. 2022. Stressed and slim or relaxed and chubby? A simultaneous assessment of gray whale body condition and hormone variability. Mar. Mammal Sci. 38(2): 801–811.
Leslie M.S., Perkins-Taylor C.M., Durban J.W., Moore M.J., Miller C.A., Chanarat P., et al. 2020. Body size data collected non-invasively from drone images indicate a morphologically distinct Chilean blue whale (Balaenoptera musculus) taxon. Endanger. Species Res. 43: 291–304.
Leslie M.S., Kant L., Perkins-Taylor C., Van Bressem M.F., Minton G., MacDonald D., et al. 2023. Remote and non-invasive quantification of ‘tattoo skin disease-like’ dermatopathy in endangered Arabian Sea humpback whales using drone photography. Mamm. Biol. [Online].
Machado A.M.S., Cantor M., 2022. A simple tool for linking photo-identification with multimedia data to track mammal behaviour. Mamm. Biol. 102(3): 961–971.
McMahon M.C., Ditmer M.A., Forester J.D. 2022. Comparing unmanned aerial systems with conventional methodology for surveying a wild white-tailed deer population. Wildl. Res. 49(1): 54–65.
Moskowitz E.E., Siegel-Richman Y.M., Hertner G., Schroeppel T. 2018. Aerial drone misadventure: a novel case of trauma resulting in ocular globe rupture. Am. J. Ophthalmol. Case Rep. 10: 35–37.
Oleksyn S., Tosetto L., Raoult V., Williamson J.E. 2021. Drone-based tracking of the fine-scale movement of a coastal stingray (Bathytoshia brevicaudata). Remote. Sens., 13(1).
Olsoy P.J., Shipley L.A., Rachlow J.L., Forbey J.S., Glenn N.F., Burgess M.A., Thornton D.H. 2018. Unmanned aerial systems measure structural habitat features for wildlife across multiple scales. Methods Ecol. Evol., 9(3): 594–604.
Perryman W.L., Lynn M.S. 1993. Idendification of geographic forms of common dolphin (Delphinus delphis) from aerial photogrammetry. Mar. Mammal Sci. 9(2): 119–137.
Piacenza S.E.H., Piacenza J.R., Faller K.J., II, Robinson N.J., Siegfried T.R. 2022. Design and fabrication of a stereo-video camera equipped unoccupied aerial vehicle for measuring sea turtles, sharks, and other marine fauna. PLoS One, 17(10): e0276382.
Pirotta V., Smith A., Ostrowski M., Russell D., Jonsen I.D., Grech A., Harcourt R. 2017. An economical custom-built drone for assessing whale health. Front. Mar. Sci. 4: 217.
Ramos E.A., Landeo-Yauri S., Castelblanco-Martínez N., Arreola M.R., Quade A.H., Rieucau G. 2022. Drone-based photogrammetry assessments of body size and body condition of Antillean manatees. Mamm. Biol. 743–757.
Raoult V., Colefax A.P., Allan B.M., Cagnazzi D., Castelblanco-Martínez N., Ierodiaconou D., et al. 2020. Operational protocols for the use of drones in marine animal research. Drones, 4(4): 1–35.
Segre P.S., Gough W.T., Roualdes E.A., Cade D.E., Czapanskiy M.F., Fahlbusch J., et al. 2022. Scaling of maneuvering performance in baleen whales: larger whales outperform expectations. J. Exp. Biol. 225(5):.
Setyawan E., Stevenson B.C., Izuan M., Constantine R., Erdmann M.V. 2022. How big is that Manta Ray? A novel and non-invasive method for measuring reef manta rays using small drones. Drones, 6(3):.
Seymour A.C., Dale J., Hammill M., Halpin P.N., Johnston D.W. 2017. Automated detection and enumeration of marine wildlife using unmanned aircraft systems (UAS) and thermal imagery. Sci. Rep. 7: 45127.
Shero M.R., Dale J., Seymour A.C., Hammill M.O., Mosnier A., Mongrain S., Johnston D.W. 2021. Tracking wildlife energy dynamics with unoccupied aircraft systems and three-dimensional photogrammetry. Methods Ecol. Evol. 12(12): 2458–2472.
Stewart J.D., Durban J.W., Knowlton A.R., Lynn M.S., Fearnbach H., Barbaro J., et al. 2021. Decreasing body lengths in North Atlantic right whales. Curr. Biol. 31(14): 3174–3179.e3.
Stewart J.D., Durban J.W., Europe H., Fearnbach H., Hamilton P.K., Knowlton A.R., et al. 2022. Larger females have more calves: influence of maternal body length on fecundity in North Atlantic right whales. Mar. Ecol. Prog. Ser. 689: 179–189.
Torres L.G., Nieukirk S.L., Lemos L., Chandler T.E. 2018. Drone up! Quantifying whale behavior from a new perspective improves observational capacity. Front. Mar. Sci. 5.
Torres W., Bierlich K.C. 2020. MorphoMetriX: a photogrammetric measurement GUI for morphometric analysis of megafauna. J. Open Source Software, 5(45): 1825–1826.
Torres L.G., Bird C.N., Rodríguez-González F., Christiansen F., Bejder L., Lemos L., et al. 2022. Range-wide comparison of gray whale body condition reveals contrasting sub-population health characteristics and vulnerability to environmental change. Front. Mar. Sci. 9(April): 1–13.
Vermeulen C., Lejeune P., Lisein J., Sawadogo P., Bouche P. 2013. Unmanned aerial survey of elephants. PLoS One, 8(2).
Vivier F., Wells R.S., Hill M.C., Yano K.M., Bradford A.L., Leunissen E.M., et al. 2023. Quantifying the age structure of free-ranging delphinid populations: testing the accuracy of unoccupied aerial system photogrammetry. Nat. Ecol. Evol. 13(6).
Wei S., Dan G., Chen H. 2016. Altitude data fusion utilising differential measurement and complementary filter. IET Sci. Measur. Technol. 10(8): 874–879.
Whitehead H., Payne R. 1978. New techniques for assessing populations of right whales without killing them. Mammals Seas, 3: 189–211.
Information & Authors
Information
Published In
Drone Systems and Applications
Volume 12 • 2024
Pages: 1 - 10
History
Received: 20 July 2023
Accepted: 4 December 2023
Accepted manuscript online: 5 December 2023
Version of record online: 11 January 2024
Copyright
© 2024 The Author(s). This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
Data Availability Statement
LidarBoX hardware documentation and build files are available through the Oregon State University Scholars Archive under a Creative Commons Attribution License. https://doi.org/10.7267/3n204680g. Details on the base system design and utilized code can be found in Dawson et al. (2017). Inexpensive Aerial Photogrammetry for Studies of Whales and Large Marine Animals. Frontiers in Marine Science 4, 366 https://doi.org/10.3389/fmars.2017.00366, https://github.com/EvaLeunissen/Whalength. Software for assisting the sync and alignment of the data logged in the LidarBoX with other auxiliary multimedia data can be found in Machado, A.M.S., and Cantor, M. A simple tool for linking photo-identification with multimedia data to track mammal behavior. Mammalian Biology 102, 983–993 (2022). https://doi.org/10.1007/s42991-021-00189-0, https://bitbucket.org/maucantor/mammals/, and https://mammals-rpackage.netlify.app.
Key Words
Authors
Author Contributions
Conceptualization: KCB, DW, CNB, RD, TC, LT, MC
Data curation: KCB, DW
Formal analysis: KCB
Funding acquisition: LT, MC
Investigation: KCB, DW, CNB, RD, TC
Methodology: KCB, CNB, RD, TC
Project administration: KCB, DW, LT, MC
Resources: KCB, DW, LT, MC
Software: DW
Supervision: LT, MC
Validation: KCB, DW, CNB, RD
Visualization: KCB
Writing – original draft: KCB, MC
Writing – review & editing: KCB, DW, CNB, RD, TC, LT, MC
Competing Interests
The authors declare that they have no competing interests.
Funding Information
The Innovation Lab, Hatfield Marine Science Center
The Marine Mammal Institute, the Marine Mammal Research Program Fund
Office of Naval Research: N00014-20-2760, N00014-23-1-2422
Marine Mammal Institute (Jungers Faculty Development and Research Fund)
Marine Studies Initiative
The development of the LidarBoX was supported by the Marine Mammal Institute's Center of Drone Excellence (CODEX), the Marine Mammal Research Program Fund at Oregon State University, and the Office of Naval Research Marine Mammals and Biology program (grant numbers N00014-20-2760 and N00014-23-1-2422). MC was supported by the Marine Mammal Institute (Jungers Faculty Development and Research Fund), the Marine Studies Initiative, and the College of Agricultural Sciences, all at Oregon State University.
Metrics & Citations
Metrics
Other Metrics
Citations
Cite As
K.C. Bierlich, Drummond Wengrove, Clara N. Bird, Robert Davidson, Todd Chandler, Leigh G. Torres, and Mauricio Cantor. 2024. LidarBoX: a 3D-printed, open-source altimeter system to improve photogrammetric accuracy for off-the-shelf drones. Drone Systems and Applications.
12: 1-10.
https://doi.org/10.1139/dsa-2023-0051
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Cited by
1. Integrated deep learning and geo-referencing for drone-based animal tracking with flexible camera angles
2. Advanced Machine Learning Models for Real-Time Drone and Bird Differentiation in Aerial Surveillance Systems
3. Innovative Machine Learning Solutions for Automated Kidney Tumor Detection in CT Imaging Through Comparative Analysis
4. Automated body length and body condition measurements of whales from drone videos for rapid assessment of population health
5. Pediatric Drone: Estimating morphometric measurements in spinner dolphin (
Stenella longirostris
) calves based on aerial photogrammetry
6. Growing into it: evidence of an ontogenetic shift in grey whale use of foraging tactics