Product Updates

To better keep our customers informed about our day-to-day engineering and manufacturing work, Lily is dedicated to providing frequent updates to the community. If you have any questions about the Lily Camera product updates below please feel free to contact us at support@lily.camera.


12/13/16

We are expanding our calibration checks to ensure the camera will function properly across a larger range of geographic areas due to the earth’s varying magnetic fields.


12/12/16

Undergoing tests to verify the accuracy of the attitude estimator’s pitch and roll by utilizing a motion capture system.


12/9/16

Enhanced video quality by slowing down Lily’s acceleration during flight commands to minimize excessive motion.


12/8/16

Improved Lily’s tracking to better handle conditions when the user accelerates out of range for prolonged periods of time. Lily Will continue following the user as much as possible at it’s maximum speed before landing.


12/7/16

Investigating relative tracking accuracy between the Lily Camera and the tracking device by analyzing the positioning of GPS satellites to determine which are being utilized by the camera and the tracking device throughout flight.


12/6/16

Improved the calibration process through enhanced magnetic field detection. If the magnetic interference is great enough to permanently alter the magnetometer, the user will be prompted to re-calibrate.


12/1/16

Undergoing extensive stress testing for the tracker’s over-the-air update capabilities by repeatedly updating between firmware builds to catch potential bugs.


11/30/16

Assessing the current magnetometer calibration process to increase reliability. This is done by identifying potential areas for improvement within the factory calibration data.


11/29/16

Identified and undergoing testing for a timing bump detection issue in which Lily’s landings can become unstable when caught at a certain angle.


11/28/16

Expanding the tracker network diagnostics to better characterize the tracker’s state of connection throughout flight.


11/24/16

Happy Thanksgiving!


11/23/16

Discovered and fixed a bug during the over-the-air update process in which Lily reaches a bad state. Now, Lily has a better error handling system to decompress over-the-air files.


11/22/16

Continued testing our automatic visual tracking of the user. This will ensure that the user is always centered within the frame.


11/21/16

Successfully verified the theoretical motor model in high altitudes and winds. As a result, Lily is able to determine the altitude and adjust the motor rpm accordingly.


11/18/16

Created a disc model to stimulate propeller characteristics while grounded to create the same current draw from the motors as flight. We can now analyze the electromagnetic affects on the magnetometer in the Lily Lab.


11/17/16

Added robustness to the tracker so that a backlog of commands requested by the user is overridden to only perform the last flight command.


11/16/16

Optimizing flight behavior by improving the tracker and Lily Camera communication.


11/15/16

Developed a theoretical motor model based on data collected from our previous environmental chamber tests. We are now testing the motor model in real world environments.


11/14/16

Finalized our onboarding tutorials to incorporate the Lily Camera states and LED feedback patterns to better educate the user prior to flight.


11/10/16

Improving autonomous landing for low battery and tracker loss at a constant velocity.


11/9/16

Improving curved path trajectories for Lily to account for radial acceleration which improves the tracking of position and velocity.


11/8/16

Utilizing motor feedback to detect potential motor failures due to propeller obstructions, debris in motors, and flight without propellers.


11/7/16

Underwent testing in cold weather conditions and found the barometer’s data becomes unreliable in temperatures below -25°C. If the temperature is below -25°C, Lily will wait for the barometer to warm up before flight.


11/4/16

Determined the theoretical maximum density altitude at which Lily is stable.


11/3/16

Testing Lily’s flight stability to determine the characteristics in extreme wind conditions and the maximum density altitude at which it can fly.


11/2/16

Changed tracker button commands to cancel a flight command by pressing any of the other three radial buttons not used to initiate the command.


11/1/16

Incorporated a software detection for underperforming SD cards during flight. Lily now auto resume records with a new video file if the previous file was interrupted and aborted to ensure that we capture your awesome moments.



10/31/16

Utilizing the motor performance findings under different atmospheric environments to measure the air density based on the power consumption of the motors.


10/28/16

Updated the tracking device’s button mapping to be more intuitive. A new double click feature has been added that functions as a “wake up” and “come to me” command.



10/27/16

Determined the motor’s operating bandwidth and the maximum thrust change possible while maintaining flight stability.


10/26/16

Creating automated calibration tests which will determine whether new firmware updates are causing any discrepancies with the calibration upload process. This will also create a test bench that can further be used for multiple feature focus tests.


10/25/16

Created a new test in controlled environment to to analyze flight stability performance during attitude oscillation.


10/19/16

Improved coverage of automated over-the-air test to catch potential Wi-Fi crashes during the update process.


10/18/16

Began testing and analyzing potential production-level camera and tracker hardware build in our manufacturing facility in China.


10/17/16

Analyzing and improving flight stability and quality for edge cases, including flights with high winds and tracking quick movements.


10/14/16

Expanding our parsing tool for flight logs to retrieve additional flight metrics. The tool will automatically analyze each log for an indepth look into Lily’s flight performance.


10/13/16

Measuring the barometer’s reaction time by utilizing our motion capture system which provides sub-millimeter accuracy to obtain ground truth data.


10/12/16

Improving state and health diagnostic visualization in our flight data management system.


10/11/16

Completed testing and implemented our improved tracker calculations for the state of charge to be more accurate. The changes were calculated by utilizing a fourth order polynomial fit.


10/10/16

Continued developing our regression tests with a controlled flight routine to determine a baseline for Lily’s flight stability. We will now be able to quantify flight control improvements using a motion capture system with each new build.


10/7/16

From our motor performance testing in environmental chambers, we were able to characterize the torque control mode behaviors to RPM control in different altitudes.


10/6/16

Undergoing extensive motor performance testing in environmental chambers to characterize Lily’s battery and electronic speed controller limitations in relation to density altitude.


10/5/16

Creating a controlled automated test environment to characterize Lily’s barometric response time during flight.


10/4/16

Enhanced the Companion App content and flow for a more intuitive experience based on feedback from our usability studies.


10/3/16

Aligning the flight controller’s desired thrust-to-RPM command based on the altitude. Higher RPM’s will be required to produce the same thrust at higher altitudes due to air thinness.


9/29/16

Utilizing a test instrument to analyze motor performance including current draw, hovering RPM, and the response time of the motors.


9/28/16

Conducting extensive life cycle testing of the propellers and motors which undergo numerous start and stop cycles to ensure the propellers properly open during their lifetime.


9/27/16

Undergoing extensive testing of the tracker’s battery level algorithm to better estimate an accurate state of charge and remaining battery life.


9/26/16

Adjusted the battery pack to better stimulate a Lily flight for our on-board fuel gauge which continuously counts coulombs entering and exiting the pack to more accurately estimate the state of charge and remaining flight time.


9/23/16

We improved the LED display of the camera eyes by changing the flashing frequency during pre-flight checks to 1.25Hz to match the display rate of the tracker.


9/22/16

Leveraged motor controller changes to measure accurate current draw per motor which allows for more detailed analysis of flight time and battery usage data.


9/21/16

Completed and implemented the model to correct accelerometer drift caused the temperature gradients within the Lily.


9/20/16

Improved user altitude following by testing the barometer’s ability to equilibrate the atmospheric pressure through the waterproof membrane.


9/19/16

Validated the waterproof membrane to ensure the barometer accurately equalizes the external and internal atmospheric pressure of the tracker. The excessive pressure caused by the compression of the buttons will dissipate through the membrane hole.


9/16/16

Creating a preliminary model to correct accelerometer drift caused by temperature gradients within the Lily.


9/15/16

The iOS Companion App reached a “feature complete” version and is currently being tested and refined to be released prior to our production launch.


9/14/16

Exploring different dewarping models to reduce the fisheye effect and create more visually pleasing footage.


9/13/16

We have completed and implemented the new tracker architecture for high priority LED patterns and vibration states. The tracker will notify the user of states including contingency hold, critically low battery, and loss of tracker connectivity.


9/12/16

Improving the robustness of communication between the tracker and Lily which results in an increase in accuracy while tracking the user.


9/9/16

Resolved an over-the-air update bug to ensure a more user friendly tracker update process.


9/8/16

Extended automatic landing to two minutes in the case of emergencies or loss of communication between Lily and the tracker.


9/7/16

Tuned the maximum rpm limits to allow for a better flight experience in dynamic conditions.


9/6/16

Adjusted the tracking algorithm offset to allow Lily to takeoff from ground from further distances.


9/2/16

Updated our calibration process by improving the geometric algorithm to ensure enough data points are taken within the fit of our mathematical model. Now, Lily’s magnetometer can better establish true north.


9/1/16

Completed and implemented the model to compensate for barometric drift over time. Lily can now adjust the temperature and pressure within the waterproof membrane for more accurate height changes.


8/31/16

Tuned the parameters of the attitude controllers which results in smoother transitions between flight modes.


8/30/16

Modified the tracking logic to ensure that the user will always be within the camera frame regardless of distance.


8/29/16

We incorporated a new feature in which Lily will power off if it stays idle on the ground or if the battery voltage is below a specific threshold.


8/26/16

We developed a preliminary model to prevent barometric drift over time caused by the temperature and air dynamics within the waterproof membrane.


8/25/16

We re-architected the tracker display to ensure the highest priority LED and vibration patterns are displayed to the user.


8/24/16

We created an automated test to measure the accuracy of the motor controller to ensure that the motor RPM’s will perform within a certain tolerance.


8/23/16

Corrected a bug that caused the GPS parameters to fluctuate which would occasionally prevent Lily from passing pre-flight checks.


8/22/16

We made the low battery readings more robust to consistently conserve more battery life which will result in safer landings.


8/18/16

Modified software to automatically power off Lily if it is accidentally turned on after twenty minutes without a signal from the tracker to save battery life.


8/17/16

Pre-flight checks for maximum altitude have been updated so Lily will not be able to pass if it is not within a safe altitude range.


8/17/16

Re-tuned the maximum RPM to allow the Lily to perform at higher altitudes and in stronger winds.


8/16/16

Revised the pre-flight check process to be more comprehensive through the Companion App for the user. Lily will be able to provide more specific reasons if a pre-flight check is unsuccessful.


8/15/16

We optimized the wifi traffic to make Lily more responsive to tracker commands.


8/12/16

Cleaned up the tracking device code to make adding future features easier.


8/11/16

Resolved a bug that offset the barometer and GPS which resulted in the two reading different heights. They now consistently determine the same height.


8/10/16

Improved calibration robustness of the magnetometer with an advanced algorithm to better manage edge cases.


8/9/16

We re-tuned the audio signal process to cope with loud noises by reducing the maximum gain.


8/8/16

Improved the GPS altitude tuning to better estimate Lily’s height which creates a more reliable flight control system.


8/5/16

We found a bug that was slowing down our tracker processor.


8/4/16

Placed a pull-up resistor that provides power to the SD card from an alternate source to resolve a bug that prevented the Lily from successfully booting up.


8/3/16

Started designing a test fixture system that connects to Lily which will automatically perform over-the-air updates.


8/2/16

Fine tuned the tracker’s microphone to enhance the quality of audio video overlay while filming.


8/1/16

Modified the tracking algorithm to try improving the camera’s zooming and framing behaviors by making it more intuitive and logical.


7/28/16

Fixed a bug in the Lily Camera’s takeoff behavior that was causing a momentary drop in thrust. Lily now changes throttle continuously and smoothly.


7/27/16

Improved the response time of the barometric pressure sensor on the tracking device by isolating it with a rubber housing. This prevents button presses from affecting the users altitude estimate.


7/26/16

Developed a process for automatic tuning that sweeps across a large number of parameter combinations and provides detailed analysis, leading to stronger height hold performance.


7/15/16-7/25/16

Over the past week, we streamlined the calibration process as a result of data and feedback we received back from our testers.


7/14/16

A new mag board was tested.


7/13/16

A characterization of the electronic speed controllers was completed.


7/12/16

LED animation bugs found on the tracking device were fixed.


7/11/16

Tuning of Lily Camera’s position estimator and controller altitude was completed.


7/8/16

Completed work to enhance Lily Camera’s metrics for offline analysis of flight data.


7/7/16

Product specs were updated to reflect the latest User Interface and flow changes.


7/6/16

Wrapped up yaw tuning and started the effort to tune Lily Camera’s position controller.


7/5/16

Preflight checks were implemented into the companion app. The user will be able to see the results on their app screen.


7/1/16

Orbit and landing speed tests were performed both in the cage and in the field.


6/30/16

A calibration rig was built by the prototyping team to streamline mag calibration testing.


6/29/16

Tutorial mode for the companion app was implemented.


6/28/16

The quality assurance team performed various tests on the lens cover.


6/27/16

Improvements made to the magnetometer were tested and evaluated.


6/24/16

A tachometer rig was built to aid with motor tests.


6/23/16

The Settings and Help sections were finalized on the companion app.


6/22/16

Improvements to the audio flow were made.


6/21/16

The Android companion app team fixed a couple of bugs found in the app’s gallery.


6/20/16

Engineering commenced a week long tuning exercise for orbit velocity.


6/17/16

We fixed a bug that was causing a motor to still spin slowly after landing.


6/16/16

A new lens cover was selected.


6/15/16

A new version of our electronic image stabilization was released with some bug fixes.


6/14/16

The prototyping team built a rig to enable the automated tuning of Lily’s attitude control.


6/13/16

The Lily Camera companion app’s navigation was finalized.


6/10/16

A change was made to the on-boarding flow. Lily Camera will be able to request that the user complete a simple calibration prior to flight.


6/9/16

Results from pressure chamber testing were reviewed.


6/8/16

Several magnetometer tests were performed for data gathering purposes.


6/7/16

We explored a few different approaches to sync the audio from the tracker with the video from the camera.


6/6/16

Flight control team focused on tuning Lily Camera’s orbit feature.


6/3/16

We made some final functional improvements to the On/Off button and its board.


6/2/16

The mobile app team tested the WiFi performance for the Gallery and to see if there is room for improvement.


6/1/16

The on-boarding video was finalized.


5/31/16

Lily Camera’s minimum position was changed to 3m above ground and 3m from the tracking device.


5/27/16

Another round of image quality review was completed.


5/26/16

Further tuning of the audio capture feature was performed. We are exploring ways to make this feature a better experience.


5/25/16

A bug that was preventing the user from reconnecting to the app after a disconnect was fixed.


5/24/16

We built a 6 meter vertical rig that will be used to help tune Lily Camera’s height.


5/23/16

Quad movement during follow mode was revised in order to have the unit travel at a more favorable angle.


5/20/16

Motor tuning improvements were implemented for faster spin-up.


5/19/16

Sonar sensor firmware integration is officially underway.


5/18/16

All visuals for pre-order packaging were locked down.


5/17/16

Several onboarding flow design elements of the Lily companion app were finalized for launch.


5/16/16

Several key stability bugs found during field testing were fixed.


5/13/16

Versions of our Android and iOS Lily companion apps were released to our Lily Beta participants.


5/12/16

We completed some research on additional ways the system could detect bad calibration.


5/11/16

We resolved an issue with the format of the data from the Tracking Device’s microphone.


5/10/16

We outlined the companion app’s interactive onboarding flow.


5/9/16

We completed the integration of computer vision into the camera framework.


5/6/16

We selected the final propellor removal tool design for production.


5/5/16

We made a UI improvement to the visual feedback the Lily Camera eyes gives the user.


5/4/16

We performed changes to the packaging and conducted new drop tests.


5/3/16

Released two different image quality tunes to compare how different colors and textures appear in the picture.


5/2/16

We made improvements to the factory calibration process.


4/29/16

Successfully made improvements to the camera cover glass.


4/28/16

Finished implementing Lily flying modes on Android operating system app.


4/27/16

Started the tooling process for the propellor guards at the factory.


4/26/16

We set up an image quality lab for characterization of the camera.


4/25/16

We implemented various user experience enhancements upon boot up.


4/22/16

Successfully brought up calibration on our manufacturing line.


4/21/16

The LED feedback on the Tracker was updated to incorporate field test findings, improving user experience.


4/20/16

We flew in various different landscapes around San Francisco for testing purposes.


4/19/16

We started accessory packaging development.


4/18/16

Parameters were finalized for the battery pack, ensuring optimal performance and lifespan for your Lily Camera.


4/15/16

We changed the priorities of threads on the tracking device to optimize performance.


4/14/16

We tuned the parameters of our camera sensor to improve image and video quality.


4/13/16

We added motor controller firmware updating to our OTA process.


4/12/16

We added extra sensor data to the log files to continuously monitor health of our field units.


4/11/16

We tuned the low battery detection to optimize flight time while maintaining safe operating conditions.


4/8/16

We conducted extensive drop and vibration testing of our packaging to prevent damage during transit.


4/7/16

We set a maximum speed relative to the user for camera change operations.


4/6/16

We tested different propeller guard samples from our manufacturer.


4/5/16

We fixed a bug that was causing jerky motion during some flight edge cases.


4/4/16

We reviewed samples of the tracking device strap and made small changes to improve manufacturability.


4/1/16

We fixed an issue with the tracking device button event handler. This was causing lockup in a few cases.


3/31/16

We developed a test rig that automatically pushes buttons on the tracking device for extensive testing.


3/30/16

We finalized the airfoil of Lily’s propellers to maximize efficiency. Dozens of designs were 3d printed, and then three versions were injection molded and tested to arrive at this final decision.


3/29/16

We developed a diagnostic to measure the health of all 4 motors.


3/28/16

We changed the opacity of the tracking device light ring to make it easier to see in bright sunlight.


3/25/16

We tuned position control gains to improve takeoff and land on ground.


3/24/16

We finalized updates to our packaging!


3/23/16

We started testing initial versions of the Lily Quick Start Guide. This will inform us if it includes enough information.


3/22/16

We incorporated Lily’s attitude estimate into our optical flow sensor. This improves the accuracy of the algorithm when Lily is tilted at high angles.


3/21/16

We changed Lily Camera’s image quality to try a few different tuning settings. We are conducting an internal survey to determine the best one.


3/18/16

We collected data from our propulsion system to create a better dynamical model of the system and improve estimation and control.


3/17/16

We fixed an issue with our message protocol that was causing the tracking device to lose ~25% of messages. This improves responsiveness of the system.


3/16/16

We developed tools to test and profile the wifi performance between the tracking device and the Lily Camera.


3/15/16

We fixed an error in our gyroscope driver. This greatly improves the quality of our attitude estimation.


3/14/16

We built an oven in which we can strap down Lily and conduct controlled thermal tests.


3/11/16

We added a sensor calibration check to the startup procedure.


3/10/16

We implemented a dynamical model of the motors to improve position control and estimation.


3/9/16

We significantly reduced the RAM usage of the flight controller.


3/8/16

We debugged an EMI issue being caused by the camera, and designed a shield to fix it.


3/7/16

We added a sensor calibration check to the pre-flight startup procedure.


3/4/16

We made some algorithmic changes to takeoff, land, and come back to user to add extra safety checks and smooth out flight.


3/3/16

We added position controller safety checks and added hysteresis on state transitions to ensure predictable behavior.


3/2/16

We fixed an issue we were having with our motors during extreme flight conditions. This involved adding one component to our motor controller boards.


3/1/16

We wrote the first draft of our quick start guide, designed our initial smartphone app run through, and worked on ensuring that the first experience of a Lily customer is awesome.


2/29/16

We improved our calibration routine to allow for a more accurate attitude estimator.


2/26/16

We made a small modification to the SD card slot to make it easier to remove.


2/25/16

We fixed a bug we identified when changing between different camera settings while in flight.


2/24/16

We boosted the speed of our optical-flow algorithm to provide better responsiveness.


2/23/16

We performed a 27 minute flight in winds of 15 mph gusting to 20 mph.


2/22/16

We added a message to allow for live streaming of sonar data to a laptop during flight. This will allow us to refine our sonar range detection algorithms.


2/19/16

We extended the web server running on Lily to allow for software updates via the smartphone app.


2/18/16

We doubled the number of LED’s on the tracking device to make the light ring easier to see in bright sunlight.


2/17/16

We spent time fine tuning height hold position.


2/16/16

We flew until the battery drained completely (27 minutes, in 15 mph winds), which successfully triggered a landing. We also took the opportunity to test throws in high wind.


2/15/16

We fixed a bug that affected low battery detection. We also added more preflight checks.


2/12/16

We fixed a bug in the tracking device that was causing the system to freeze. This turned out to be the result of our GPS thread having too high of a priority and context switching at bad times.


2/11/16

We adjusted pan and zoom flight parameters. This will make for better cinematography.


2/10/16

We integrated our optical flow camera into flight navigation and performed fully gps denied position hold.


2/9/16

We developed pan and orbit features on top of our trajectory generation system. This allows Lily to smoothly rotate around the user while in motion.


2/8/16

We conducted a 24 min endurance flight at 7000 feet elevation. All components of Lily operated without issue.


2/5/16

We developed a computer vision algorithm to detect land in hand using the bottom facing camera. This, combined with bump detection using the accelerometer, allows us to very reliably detect contact with a user’s hand.


2/4/16

We wrote a trajectory generation algorithm to create circular paths around the user. This generates position, velocity and acceleration targets for Lily.


2/3/16

We reworked a tracking device to evaluate the quality of an alternative microphone. We experimented with different style LED’s on the tracking device to boost brightness. We tested another design of our propeller guards.


2/2/16

We developed an automatic rig to test the optical flow camera on the bottom of Lily. This takes the camera along a known path and then compares the estimated movement with reality.


2/1/16

We finished running through our standard bench and field tests for the latests units off the assembly line. Unmodified units fly beautifully with our production motor controllers.


1/29/16

We implemented a timeout safety feature on our motor controllers. After over a week of investigation, we fixed a problem in one of our I2C drivers that was caused by a bug in the silicon of the chip.


1/28/16

We developed smoother zoom functionality using trajectory generation. We implemented commands for high level control over via the smartphone. We developed an HTTP web server so that videos can be downloaded from Lily via Wifi.


1/27/16

We received another revision of our hardware and began testing it. Initial flight results show very stable and reliable flight. We investigated a timing problem in the motor controller driver and narrowed down its cause. We tried another potential design of the propeller guards.


1/26/16

We began a refactor of some of our state transition logic to better integrate trajectory generation. We did a review of all hardware changes in our most recent revision of the hardware.


1/25/16

We completed the development of Lily’s height following feature that matches the altitude above the user at all times. We fixed a software bug that was causing our IMU to stop working after extended usage.


1/22/16

We successfully tested throw in the air takeoff with our new motor controllers. This required us to reduce the time it takes to spin up from 0 rpm. We finished the first version of our over-the-air update system. This allows us to improve software for all units for the lifetime of the product.


1/21/16

We conducted tests on a new propeller improvement. Initial results show improved efficiency and maximum thrust. We conducted tests on an improvement to our motor controllers. The flight characteristics of these controllers are very smooth and responsive.


1/20/16

We finished setting up a cloud based build system for all of our code. This allows us to make regular stable releases of our system for testing. We conducted many tests on our battery to determine shelf life after Lily has been fully discharged.


1/19/16

We tested a variety of methods for solving the I2C motor controller problem. Nothing has fully solved the issue yet. Started testing the production iOS app with our production units. Developed first draft of an automated build system that compiles all binaries for all components in the system.


1/18/16

We dug deeper into an I2C communications problem with the motor controllers. We figured out that it is caused by another thread context switching during an I2C transaction. We did more extensive and aggressive throw testing.


1/15/16

We fixed a bug in our GPS driver on the tracking device. This improved the data rate significantly and made user tracking more reliable. We finalized the lengths of all charging cables to be shipped with Lily.


1/14/16

We added functionality to change the bitrate of the video recording and experimented with different settings. We added a lot more functionality to our debugging smartphone app. We changed the mapping of inputs to thrust on our motor controllers.


1/13/16

We removed the need to use a debugging tool during development for easier field testing. We boosted the maximum thrust output of our motor controller system. We tested and tuned our trajectory generation system for more responsive and accurate flight paths.


1/12/16

We spent more time tuning Lily’s new motor controllers. We added smoothing to Lily’s following paths to ensure better video quality. We investigated changing GPS tracking parameters to achieve more accurate following.


1/11/16

We did the initial integration of trajectory generation into the flight system. We added functionality to log flight data on the camera system. Updated the tracking device strap to prepare for tooling release. Modification of charging cable for tracking device.


1/8/16

We collected battery discharge data to better characterize flight time. We piped the data into a programmable test load to reproduce flight profiles on the bench. We added more status messages to the debug smartphone app. Fixed a bug with velocity estimation on the tracking device.


1/7/16

We added a reliability layer to the tracking device button signals. We modified our debug and development smartphone app to allow for easier field testing. We added log collection to the new integrated system to allow for easier debugging.


1/6/16

We upgraded the provisioning system on the Lily Camera to change the SSID and password of the access point. We wrote the initial implementation of a trajectory generation system for smoother flight paths.


1/5/16

We tested throw takeoff outside. Looking great! We fixed a bug with recording that was causing video files to stop half way through a flight. We profiled our new motor controllers and tuned them for responsiveness. We finished setting up our over-the-air update system for the tracking device.


1/4/16

We implemented throw takeoff and tested it in our indoor controlled environment. We finished modifying our new propeller design and tested it on a prototype.


12/31/15

We released all the PCB boards for the next pre-production build. We changed some wiring to reduce sensor noise and vibration. We also made the Lily Camera more operator-friendly for assembly.


12/30/15

We tested the power adaptor prototypes and they are working well. We are improving a few cosmetic issues and will review another round of samples in a couple weeks.


12/29/15

We assembled a few lens and image sensor modules to tune the optical center alignment. They are on the way to Lily US for testing.


12/28/15

We are getting samples of different material for the tracking device strap. We are discussing with the factory about different types of tooling and making design modifications to prevent breakage when taking the parts out of the mold.


12/24/15

We tested an improved design of the tracker light guide to address deformation issues during the molding process. The new light guide diffuses light better and still passes for waterproof so far. We will continue to run other mechanical tests.


12/23/15

We finalized the state machine for the flight controller, taking inputs from the camera, tracking device, and smartphone. We started testing different vibration dampener durometers to improve video quality. We set up a debugging system to send data through the camera system onto a desktop.


12/22/15

We integrated the camera system into the flight controller to allow for message passing. We finalized all button combinations and LED patterns for the tracking device. We tested more efficient propellers.


12/21/15

We integrated the tracking device GPS signal and tested the full feature set with the full DVT system. The tracking device position and velocity estimate is looking very responsive and accurate. We also added more button and LED functionality to the tracking device.


12/18/15

We tuned the control loop running on our new motor controllers and achieved stable flight. We refined the design for our new propellers and prepared to move to final tooling. We fixed a few power bugs on the tracking device and added more status LED messages.


12/17/15

We ran an endurance test on a DVT unit flying continuously for 21 minutes and 18 seconds without any problems. We collected data to characterize our motor controllers more precisely. We integrated the DVT tracking device with the DVT camera and will test the full integrated system tomorrow. We finished building the first version of a development smartphone app to be used to debug problems in the field. We further refined our propeller guard designs to make them more securely mounted onto Lily. We started modifying our more efficient propeller model to fit the folding design.


12/16/15

We tuned our motor controllers to fly with more stability. We also modified the boot up and shutdown sequence on the tracking device to be more easily operated on battery. In addition, we set up our indoor testing system to mimic the tracking device for evaluation purposes and fixed a driver level bug in our optical flow sensor – it is now fully functional.


12/15/15

We flew the DVT units for 18 min in our testing area. Voltage and temperature still held up and we could have flown for a few more minutes. Battery life is looking good (even without the upcoming weight reductions)! We tuned our motor controllers and greatly improved the stability. We fixed a diode problem in our optical flow board.


12/14/15

We set up a magnetometer calibration routine and took the DVT units out to fly fully autonomously. We tested takeoff and land in hand, watch user mode, and follow user mode. We wrote a program that allows to update the firmware on our motor controllers over the air. We finished bringing up the DVT tracking device, and will take it out to test tomorrow.


12/11/15

The DVT units can now do takeoff and land in hand! We also collected video from these units while in flight. The image is smooth and does not show any rolling shutter artifacts in our indoor clips. Next, we will be testing the images in outdoor light to determine if the brighter environment will cause any problems with image quality. We also finished setting up the motion capture system and were able to fly fully autonomously using this data instead of outdoor GPS data.


12/10/15

The DVT units are now flying fully functionally! We swapped out the new DVT motor controllers with our old prototype controllers to isolate the problem. With the old prototype controllers, the unit flies perfectly! We brought up the DVT tracking devices but there was a pin change for the GPS module so we need to rewrite a few pieces of our code.


12/9/15

We started collecting temperature data from the battery in the DVT unit with the motors running at full flight speed. We also built a test Jig to more easily program and debug the DVT tracking device. In addition, we collected data from the DVT magnetometer so that we can design a calibration routine.


12/8/15

We wrote a driver to measure the battery voltage, current and temperature. This, combined with temperature measurements from our in-arm motor controllers, will allow us to profile the thermal characteristics of Lily Camera. We also developed a system to more easily compile our code to run on the DVT production units as well as our development hardware.


12/7/15

We fully brought up the new DVT unit and were able to begin testing flight. The motor controllers have not been fully tuned, however, and the flight is not yet stable. We collected height data so that we can begin tuning the altitude controller next week.


12/4/15

We received 6 more DVT units and modified them for development. Some of the motors are displaying technical issues when running without a propeller attached, but we have identified the possible causes and are working through the problem. We finished setting up the motion capture system to characterize the performance of Lily Camera’s flight.


12/3/15

We brought up everything on the first DVT unit but had issues with one of the motors on the unit. We are receiving another 4 units from the factory tomorrow and will then be able to fly with full systems. We built an area with 20ft high netting for indoor testing.


12/2/15

Lily Camera is now successfully flying indoors with just the downward facing camera (no GPS signal). We received the first fully assembled DVT units from the factory and were able to reach the halfway point in flight testing set-up.


12/1/15

We successfully tested all sensors on the DVT build of our electronics. We also set up a motion capture system to better analyze Lily Camera’s flight characteristics.

Banner-Lily-ProductUpdates_2016-04-29
Thank You!

We will get in touch soon.

Get Your Lily Camera Now For $499!

(12 Days To Go) Shipping in February 2016.
Lily with Tracking Device and Case $999.00 $499.00 x
(12 days left at this price!)
Payment
Account
Subtotal 499.00
Shipping & Handling: 8.00
Sales Tax (8.8%): 35.10
Total: $546.92
* You will be charged immediately. Taxes are based on a maximum US tax tate and excess tax will be refunded based on your shipping address.

Powered by Stripe
Secured with 128-bit encryption
By clicking “Pre-order” you agree to the Terms and Conditions.