GPS Backpack Herbicide Tracker

Motivation

Invasive plant management is a big part of conservation work. Nature preserves, state and federal parks all need to manage and spray invasive plants. This is mostly done with backpack sprayers or ATVs, and is a physically demanding job. However it is important as invasive plants cause huge problems to native ecosystems. They outcompete native plants and can even cause extinctions if not kept in check. It is for this reason it is important for land managers to know where they have worked in the past and be able to quantify their work to better inform them on what to expect in the future. Current methods that I have used to do this are with Garmin GPS devices and cell phones. But these products do not seamlessly collect data on where herbicide has been applied and to do so would require workers to remove their gloves and take points manually which is hard to do. This is why I have designed and built a device to collect herbicide sprayer data that does this automatically.

Intro

This device is an Arduino Mega in a waterproof box that records the applicator's GPS tracklog and spray data. It mounts to the metal stand at the bottom of the backpack and has a magnetic switch placed in the sprayer handle. If the handle is briefly pressed, it records a single GPS point, and if the handle is continuously held, it records a series of points. The data is saved into two CSV files, which can be imported into any GIS software where markers can be created around the GPS points and hotspot analysis can be performed. The photos demonstrate how it mounts and the kinds of data it collects.

I believe this device is a helpful alternative to other spray data collection methods. I was inspired to build it during an Invasive Plant Management job I worked at last summer. I found it very clumsy trying to collected GPS data manually using a handheld GPS or iPad. By comparison, once my device is installed and turned on, the sprayer can be used exactly as normal, no secondary operations or button presses.

I have wanted to create this device for several years; I have done invasive plant management for several non-profits and state parks. Once collected, the data from this device allows us to track many useful metrics. Is the herbicide creating benefits year after year? What is the minimum dose of herbicide to see benefits within a given season? How effective is a given herbicide in different habitat types?

While I designed the data logger to fit cleanly in a backpack sprayer, it could easily be installed to fit most other types of herbicide applicators. My hope is that by making this device open source and by nsnnsnnnnoffering a step by step build tutorial other people will be inspired to participate in these studies.

It was my original intent to provide very precise precise drill and cut instructions. After building 10+ of these devices I've discovered almost every box needed tiny tweaks to hole size and location. I therefore thought it would be more helpful to provide a framework of drill and cut steps to be adapted to the specific enclosures, switches and other odds and ends you may purchase yourself, rather than give rigidly prescriptive instructions.

Drill a hole in the bottom of the box and cut the lattice to create clearance for the power switch

Add a hole for the strain relief connector. This can be drilled in the middle or the left side of this face of the enclosure. This is the hole that will pass the hall effect sensor's wire.

Drill or cut out screw boss to create clearance. Note that the blue mark indicates the boss that needs to be removed. We needed to remove the screw boss next to the power switch to make room for the Velcro that holds the battery.

(Note in the provided picture the rocker switch is shown as already installed, we recommend doing this at a later step)

Place Velcro to secure battery (not pictured) and zip tie the Arduino and GPS shield to the lattice.

Near the bottom of the picture, note the screw position changed due to the original whole being covered by the Arduino

Zip ties should be ⅛ in wide

Attach wires to the power switch. As pictured Red is power from the battery, yellow is power to the arduino, and blue is ground. Note you will need 2 ground wires, one to the Arduino and one to the battery.

Install rocker switch in box

Attach red wire to battery, yellow to VCC, and ground to Arduino ground

Solder switch wires to Arduino. Brown to 5v, blue to ground and black to digital 11.

We found it best to mount the Hall effect sensor using a hose clamp and a spacer that we 3D printed.

The 3D print is convenient but is nothing fancy. If you do not have a 3D printer, you can craft a spacer out of rubber or plastic and wrap it around the handle then clamp the sensor in place.

Determine + and - for the magnet and appropriately place on the handle with glue and shrink tube. You will want to test that the magnet is at the right distance from the hall effect sensor before fully committing to the magnet location. The hall effect sensor we used has an LED indicator to ease this debug step. Zip tie the hall effect wire to the tubing.

Place the GPS antenna with 3M adhesive. We found that above the Arduino is most convenient. The Adafruit GPS shield comes with a passive GPS module, but I think the Active module does a better job. You can test it without purchasing the external GPS to see if you need higher accuracy.

Measure, cut and bend the trestle to secure the box to the metal bar of the sprayer. Also, place clamps like in the photo.

Make sure you are using charged batteries. Flip the power switch to ON. The GPS shield has an LED that flashes an indication pattern to signify when a GPS connection has been established. We found it takes a minute or two at most.

Use the sprayer as normal. When you squeeze the sprayer handle the magnet will move close to the hall effect sensor telling the logger to record a data point. After spraying herbicide on all the pesky invasive species make sure to turn the device off to preserve the batteries.

Remove the SD card and check out your data!

Here are tutorials on how to process data in QGIS.

https://www.youtube.com/watch?v=RU8xiigib3Q

https://www.youtube.com/watch?v=fDxJds9HLB0

https://www.youtube.com/watch?v=J7QQxEOX1kQ