This thread will be to organize discussion on the concept of a multipurpose multirotor utility drone that could accept multiple payloads to accomplish different missions.
I believe that this is a project that fits Arrow for several reasons.
A cheaper BOM will allow Arrow to more quickly build the same design across many different workshops and the lower safety standard we’ll have by avoiding manned flight will allow us to move more quickly on each iteration. Additionally, we should have a faster path to actually selling this product to customers with this drone vs Feather. This will allow Arrow to experiment with revenue flows back into the DAO treasury and it will give us a financial flow that we can improve through iteration.
The more affordable design, combined with a modular attachment system, is intended to allow us to better focus on growing Arrow’s community and improving our open collaboration culture. If we build a robust base drone, our community may become a natural gathering point for anyone looking to perform utility work with a drone. Ideally, we could see a growing catalog of community-built attachments for this drone.
In order to make our best attempt at setting requirements for the base drone, we can crowdsource some of our ideas for attachments and missions of the drone. Please contribute any ideas you have! If relevant, include things like:
Accessory estimated weight
Mounting location on the drone
Integration requirements (power, data, etc)
Flight time required for the mission
Various mission details (range, altitude, special considerations)
Additionally, please share any general requirements that you think may be worth considering for the drone.
We can be quick with this process and not let perfect be the enemy of progress. Using the data we gather here, we can collectively set and freeze requirements for our first prototype next week and get to work.
In terms of Interface and Controls, I think a UX with a Tablet + Joystick would be sensible way to go.
We could use a device like this:
This way we can combine both the high level mission planning and low level override control in the same compact unit.
If we will be using a pixhawk or pixhawk compatible flight stack, then the tablet can access Q ground control or Mission planner (droid planner) directly via their android apps. No need for any more UI dev
The customer could also then use their own IOS or android devices meaning reduced cost.
Fire retardant deployment - could be most effective and needed near populated areas, (cost of drones more justified due to potential loss of life and economic impact of fire) deployed in a swarm to contain a fire as quickly as possible.
Water dropping for fire suppression - Could be strategically placed near reservoirs or other large bodies of water with pre planned water (scooping?) positions for continuous work without interruption. Small crews could be manning battery replacement landing zones
Fire reconnaissance/tracking
Emergency evacuation announcements - live updates of evacuation routes that are safe and which ones are not via app or emergency notifications
Larger aircraft positioning and guidance
Delivery of first aid, equipment, for hard to reach personnel
Marking potential fire lines, tree/brush removal for on ground personnel
Search and rescue assistance - delivery of first aid/food and water to those in need
Initial costs of this system would be higher than traditional fire fighting methods, but would pay off in the long term. - The more these systems are developed the cheaper they will be in theory
Drones are much safer, cheaper, and can reach more difficult areas than traditional aircraft such as tankers or helicopters
Drones are more precise than tankers and multiple can be used together to coordinate more precise containment and help direct fires more effectively
*Will add more as I conduct more research
Concerns brought up during Engineering call:
-Larger drones might not be as effective as a smaller drone swarm: This is a possibility but I will have to read more into it.
-Larger aircraft such as helicopters and tankers are more effective: They are more effective for longer distances and carrying capacity, but for rapid response, precise movements, coordination, they are potentially more effective.
-Larger drones are more expensive to maintain than smaller ones: I would want to explore if larger drones are more expensive to maintain than helicopters, this would potentially make the case for replacing them for the purposes they serve in fire fighting.
Here’s a list of potential attachment ideas/missions that I’ve had in mind:
Agricultural herbicide dispenser. Brushbullet seems like an interesting product that neatly fits the application we have and they already have a dispenser that could be mounted. In West TX, there is a need for using targeted herbicide to help with native grassland restoration efforts, however we may have ethical/ecological concerns about herbicide dispensing for other purposes. This could be mounted underneath or on the side, however it likely needs to be loaded from the top. Does not require high power or data beyond a relay.
Extra battery attachment. Could be used to increase endurance when the primary attachment is a lower weight.
Communications relay. Could be used for repeating radio signals in disaster response areas without line of sight or for an instant cell tower after a natural disaster.
Cargo box. Simple enclosed box to transport small bits of cargo.
Tree trimming. Benjamin mentioned this originally - a small chainsaw could be used in combination with a robot arm to trim trees.
Tethered ground power. We could send power up from a ground unit for applications when high endurance is required, but it doesn’t need to fly any distance. Pairs well with communications relay and flood light attachments.Could partially replace the battery.
Loudspeaker.
Computer Vision/ML computer for dev work or visual positioning without GPS\
Ground station where the drone can land and automatically be charged/battery swap
Possible Requirements/design specs:
Batteries that can be easily swapped without any tools
Collapsible motor arms/landing gear
Option for various lengths of landing gear
Built-in basic fpv camera for navigation
Radar/Ultrasonic/Vision-based collision avoidance
Add on generator for testing as a hybrid powertrain
Questions and thoughts:
Power and data interface. Do we want CAN only, or a simple relay also? We could make a separate simple CAN to relay adapter. Ethernet/PoE could also be interesting to consider.
We should set a firm completion date for the first version. I don’t want to spend a ton of time designing a perfect base drone, rather have quick iteration cycles.
Do we want the ability for multiple attachments to be used at the same time?
Are there multiple different mounting locations for attachments?
We could design a pelican case or something similar for easy and secure transportation.
We could design a charging case that lets multiple batteries charge in a case for easy quick swapping operations. (See DJI battery station)
