Thursday, December 24, 2015



This paper is based on the REMUS 600 autonomous underwater vehicle (AUV) designed by the Oceanographic Systems Laboratory of the Woods Hole Oceanographic Institution. This 12.75 inch in diameter AUV was developed for variety of underwater missions including environmental research, underwater mapping, and performing mine countermeasures operations.

The REMUS 600 features a modular design, which allows for easily reconfiguring of its sensors for different missions. The vehicles endurance is up to 70 hours at speeds of 5 knots. The REMUS 600 can operate at depths up to 600 meters. With its increased payload it has a range of 286 nautical miles (Patterson, 2009). 

The vehicle is equipped with variety of sensors, which include a Kearfott KN-4902 Inertial Navigation System (INS), and an Acoustic Doppler Current Profiler (ADCP). A tail mounted GPS antenna also provides WIFI connectivity at up a 2.5 miles range. An Iridium satellite link is used for long range communications with the vehicle, enabling the AUV to “call home” from virtually any location. The REMUS 600 also carries a micro modem supporting the Compact Control Language. This feature allows real-time transmission of acoustic vehicle parameters and data to either a laptop or the vehicle tracking device. This control feature allows the operator to follow the status of the AUV and monitor the mission progress.

Other sensors installed on the AUV include pressure sensor, side scan sonar, temperature, and conductivity probes. Optional sensors include a dual frequency side scan sonar, a video camera, an acoustic imaging sensor, a synthetic aperture sonar, an acoustic modem, and a fluorometers (Kongsberg, n.d.).

These sensors provide data on the vehicle internal status and its location as well as a variety of environmental data. Such information as water quality, salinity, fluorescence, temperature, and bathymetry. The synthetic aperture sonar produces high resolution images combined with a large swath width as compared to conventional single beam side-scans. Laser The Scaler Gradiometer (LSG)/ Re-acquisition Payload assembly combines a magnetometer with an electronic still camera and short range, dual frequency side scan sonar. It collects data from a large search area and provides classification of targets such as buried mines. The sensor data is downloaded at the end of each mission upon recovery of the AUV (Stokey & Roup, n.d.).

By joining the sensor data with the navigation data, 2D and 3D visualization of the environmental parameters can be displayed. Data presentation methods range from color coded diagrams, still images acquired by the camera, and navigation maps from sonar imagery taken by side scan sonar. The temperature sensor data can be presented via diagrams with color coded information and combined with depth sensor and navigation data to present a complete picture of the area. Figure 1 presents the data presentation methods in a form of diagram. Figures 2 and 3 present the data from the camera and side scan sonar.

Figure 1. Water temperature at various depths. Adapted fromDevelopment of REMUS 600 AUV,” by Stokey, R., & Roup, A. (n.d.). Copyright by WHOI.

 


  

Figure 2. Data presentation methods: 5 meter altitude electronic still image, with 200 W-S strobe illumination (right) and 900 kHz, 30 meter range scale sonar image (left). Adapted from Adapted fromDevelopment of REMUS 600 AUV,” by Stokey, R., & Roup, A. (n.d.). Copyright by WHOI.




Figure 3. SAMS II, MSN012, 300kHz Side Scan Image showing transit across fault. Adapted from Adapted fromDevelopment of REMUS 600 AUV,” by Stokey, R., & Roup, A. (n.d.). Copyright by WHOI.
 
The control station of the REMUS AUV can be based on shore or on a support ship. The operator is interacting with the vehicle using simple laptop. The AUV control computer is based around the PC-104 technology. The CPU is assembled on a motherboard with eight 12-bit analog to digital channels, input/output ports, power supplies, and other interface circuitry. The user controls the AUV and monitors its status through its diagnostic software and communicates via an RS-232 serial link. The REMUS vehicle interface program, called REMUS VIP is designed to run on a laptop equipped with Windows. The Vehicle Interface Program (VIP) is used for control and monitoring of the AUV. The interface program allows the operator to plan the mission, perform vehicle operational maintenance checks, download and analyze sensor data. The VIP features a Windows based operational interface, a quality control check, and an easy to read vehicle parameters indicators ("REMUS," n.d.).

The vehicle diagnostic software display the status of all vehicle sensors. The operator can easily control the vehicle via sliders and buttons. The mission route can be uploaded via the VIP interface, which includes route points and the location of the transponders. As we can see from the Figure 4 the status of all major sub-systems and diagnostic messages is available to view on a single screen, which reduces the need to switch between screens and allows the operator to focus his attention on one monitor. All telemetry data is recorded by the vehicle during its mission and is available for later review. This interface design allows the operator to control the vehicle while monitoring vehicle health and sensor data at the same time (Woods Hole Oceanographic Institution [WHOI], 2007).



Figure 4. Graphical user interface is highly intuitive. Adapted from Adapted from Adapted from “Development of REMUS 600 AUV,” by Stokey, R., & Roup, A. (n.d.). Copyright by WHOI.

 Communication between the vehicle and the operator is conducted via a standard Ethernet connection. A graphic user interface is intuitive and allows the controller to view a map of the mission at any time, it includes an integrated text editor for uploading the mission parameters. The software also features an automatic error inspection performed on all aspects of the planned mission. The error check generates a warning messages in case any mission parameters are incorrect. A color coded messages of mission parameters allow for at a glance vehicle check: green color indicates normal status, and red indicates a fault.

 The REMUS AUV can also use gateway buoy for communication with the control station. It allows for remote monitoring, tracking, and control. The AUV may surface, uplink with the buoy, and send information to the operator.

 One of the challenges of REMUS control pertain to the vehicle’s operational environment. The underwater operational domain present such restrictions as control and datalink signal constraints. GPS, Iridium, and WiFi signals are unavailable when AUV is operating under water. Retrieval of the mission data is performed after the vehicle surfaces. Most of the information from the vehicle needs to be processed after retrieval. One of the suggestions for improving AUV operation is installing data processing system directly onboard of the vehicle, therefore data could be preprocessed before it is downloaded. When the AUV surfaces, the processed data can be send over limited bandwidth to the operator. It can help review the mission in real time and make decisions on how to proceed with the survey in the most efficient way. If no bandwidth is available, the preprocessed data can be quickly uploaded at the end of the mission and will require minimum processing after upload.

Another improvement that can be recommended pertain to the control station design. Most of the data from the REMUS is presented in a visual format, which is a one-sided approach. Although the computer screen presents a variety of useful sensory information for the controller, it can be overwhelming and fatiguing. Although the control station feature a simple interface based on a windows laptop, some improvements may help increase operators situational awareness. For example, incorporation of audible alarms in case of vehicle parameters are out of normal. Although most of the AUV operation is autonomous, while vehicle is on the surface it is possible to control it path by the use of a control joystick that could incorporated a haptic feedback system.


References





Monday, December 7, 2015

Draganflyer X8 data format, protocol, and storage


Data gathered by the variety of sensors installed on the unmanned aerial vehicle (UAV) is an essential component of the vehicle’s mission. The methods of data storage, processing, and dissemination can differ from one system to another. This paper focuses on the Draganflyer X8 helicopter UAV used for aerial video and photography. This UAV performs a variety of missions including wildfire control, law enforcement operations, crime scene investigations, industrial inspections, and commercial photography and videography. The Draganflyer features a quadcopter design which allows the vehicle to ”hover and stare.” The UAV has eight blushless electric motors which allow for heavier sensor payload capability and more flight stability even in the windy conditions. The Draganflyer X8 can carry up to 1.7 pound in camera equipment for photo/video and surveillance missions and can fly for up to 20 minutes on one battery charge (Draganfly, n.d.).

The UAV’s sensors suit includes camera equipment and a total of eleven sensors which are responsible for vehicle position and control. The standard camera sensor is a Sony NEX5R with an optional thermal FLIR (Forward Looking Infra-Red) attachment. The camera features a remote tilt, shutter, and a wireless video feed (FLIR Tau 320 9Hz). The video recorder also has onboard data storage. As an option, the Draganflyer can be equipped with a remotely operated 10MP still camera and a 1080p video camera ("Draganflyer X8," n.d.). Figure 1 depicts the Draganflyer X8 and points out some of its key features.


Figure 1. Draganflyer X8 and its sensors and features. Adapted from “Draganflyer X8 tech specs,” by Draganfly, n.d. Retrieved from http://www.draganfly.com/uav-helicopter/draganflyer-x8/specifications/ Copyright by Draganflyer.

An additional upgraded camera sensor for the Draganflyer X8 is the new IP video camera. The IP video cameras distribute digital video via an 802.11n 5.8GHz Wi-Fi connection. The advantage of digital video is that it is less susceptible to random noise than standard analog video. The captured picture is changed to zeros and ones which in turn are transferred as high and low power radio signals. The receiver changes these low and high power signals back to zeros and ones where a computer reconstructs it back into a video. This digital data transmission method eliminates the negative effects of signal noise. The video is transmitted over a dedicated wireless network and is encrypted for additional security. This also safeguards the data from unauthorized viewing ("Digital and raw data," n.d.). The digital video can be stored in the internal memory before transmitting it to the ground station. Also a real-time video stream can be transmitted over the internet to the end user.

The data format produced by the Draganfly varies depending on its mission. For commercial photo missions the digital camera records images in the most common formats such as JPG, BMP, and TIFF. Basically, when a picture is transferred from a camera, it’s coded as different lighting and color shades for each pixel on the camera’s charge coupled device (CCD) device. The CCD chip consists of an array of light sensitive pixels. Each pixel generates an electric current when a photon strikes it, this is known as the photoelectric effect. This current is read from each pixel and then recorded in memory as a series of light levels and colors as a raw image. After that raw format is converted into JPG, TIFF, or BMP format.

However, for law enforcement applications the raw file format is the best form of data used as evidence. By using the raw file format in the investigation law enforcement personnel are less likely to miss any detail, which may be lost during image processing.

The ground control station (GCS) connects to the UAV via a 2.4GHz control, sensor, and adjustment link with the Dragan Eye Pro 5.8GHz wireless video receiver. The video feed from the vehicle can be viewed in real time either on the monitor or using video glasses. The handheld GCS is optional, but it features an innovative design by combining the handheld controller and the video terminal in one compact station. It runs on a Linux operating system on an Intel Atom processor. The operator can fly the vehicle, view the video feed form the camera, and monitor UAV status on the same display.  Figure 2 represents split view screen for video monitoring and the flight parameters display.


Figure 2. Flight parameters and video screen. Adapted from “Draganflyer X8 tech specs,” by Draganfly, n.d. Retrieved from http://www.draganfly.com/uav-helicopter/draganflyer-x8/specifications/ Copyright by Draganflyer.

 

Several sensors are installed on the X8, which support the vehicle’s navigation, control, and autonomy levels. These sensors include: 3 gyros, 3 magnetometers, 3 accelerometers, an inertial measurement unit (IMU), a barometric pressure sensor, and a GPS receiver. The pilot can choose the vehicle’s degree of autonomy, which include altitude hold, position hold or manual throttle. The operator can select manual throttle mode during which he can control vehicle’s altitude by use of thrust and the UAV will automatically holds the selected heading and maintain level flight using the IMU and magnetometer. A preset altitude can be also automatically controlled using the pressure sensor. Automatic position hold is accomplished by using a combination of the pressure sensor and the GPS (Nahon, Sharf, Harmat, & Khan, n.d.).

The GPS sensor is a nice addition to the sensor suit of the UAV. The GPS is used for vehicle tracking, navigation, and position hold. The GPS has a backup power supply Lithium polymer battery, providing redundancy in case the main power source is depleted. It also ensures that vehicle location and speed data will be accessible to the operator even in case of main power failure. The GPS position update rate is 4Hz. All the GPS data is being uploaded to the “black box” recorder on the Draganflyer.

The “black box” flight recorder feature becomes handy for post flight study, crash log reviews, or malfunctions troubleshooting. The “black box” consists of removable 2GB MicroSD memory card installed onboard of the UAV. It loges data from onboard sensors and records such parameters as datalink quality, speed, orientation, altitude, and heading.

Power to the vehicle’s motors and sensors is supplied by the rechargeable 14.8V Lithium Polymer battery with 5400mAh capacity ("The Draganfly technical reference manual," n.d.). Voltage requirements for the camera sensor is 8 Volts to 32 Volts and power consumption is less than 2 Watts. The Dragan Eye 5.8GHz Wireless Video Transmitter is regulated at 5V DC and capable of sourcing 0.8A (RCtoys, n.d.). Barometric pressure sensor and airspeed sensor’s power requirements are between 3V and 16V. For the accelerometers, power requirements range between 4V to 16V for low G sensor, and between 4 to 6 V for high G sensor ("Draganfly technical reference manual," n.d.).

The Draganflyer X8 features a low battery power warning system, which monitors the battery voltage and transmits voltage information to the operator control station. Both audible and visual warnings will alert pilot of a low power situation. This alert allow enough time for the operator to land the vehicle safely. The GPS position will be recorded for vehicle retrieval.

The UAV uses two different frequency bands. One for its video sensor data downlink and separate one for the control signals. This technique reduces interference between the two frequency bands. The control signal is broadcasted at 2.4GHz and the video signal is transmitted at 5.8GHz.

The author recommends a possible improvement for the UAV’s data handling and treatment. It is possible to use the cloud architecture for data processing and distribution. By downloading real-time video and photo data to cloud, multiple users can have access to it simultaneously. It would be especially beneficial for law enforcement, crowd control, reconnaissance, search and rescue, and wildfire control. The Cloud architecture would provide secure data storage, and easy fast access to the necessary data for the end user. The Cloud architecture would also be useful if several UAVs were used to perform a certain mission at the same time. For example, if a formation of the UAVs were used to survey the large area, uploading the video data to the cloud would allow the user to view the “bigger picture” in real time with fast access.


References