DEPRECATION NOTICE: This repository has been deprecated in favor of the next generation of Shepherd.
Batteryless sensor nodes depend on harvesting energy from their environment. Developing solutions involving groups of batteryless nodes requires a tool to analyze, understand and replicate spatio-temporal harvesting conditions. shepherd is a testbed for the batteryless Internet of Things, allowing to record harvesting conditions at multiple points in space over time. The recorded data can be replayed to attached wireless sensor nodes, examining their behaviour under the constraints of spatio-temporal energy availability.
Features
- High-speed, high resolution current and voltage sensing
- Technology-agnostic: Currently, solar and kinetic energy harvesting are supported
- Remote programming/debugging of ARM Cortex-M MCUs using Serial-Wire-Debug
- High resolution, synchronized GPIO tracing
- Configurable, constant voltage power supply for attached sensor nodes
- Level-translated serial connection to the attached sensor nodes
For a detailed description see our Paper.
A shepherd instance consists of a group of spatially distributed shepherd nodes that are time-synchronized with each other. Each shepherd node consists of a BeagleBone, the shepherd cape and a particular choice of capelets according to the user requirements.
This repository contains the hardware design files for the shepherd cape and the various capelets, the software running on each shepherd node as well as the tool to orchestrate a group of shepherd nodes connected to a network.
Start by assembling your shepherd nodes, consisting of a BeagleBone Green/Black, a shepherd cape, a harvesting capelet and a target capelet. The next step is to manually install the latest Ubuntu Linux on each BeagleBone. You can install it to SD-card or the on-board EMMC flash, following the official instructions. Make sure to follow the instructions for BeagleBone.
The following instructions describe how to install the shepherd software on a group of shepherd nodes connected to an Ethernet network. We assume that your local machine is connected to the same network, that the nodes have internet access and that you know the IP address of each node.
If you haven't done it yet, clone this repository to your local machine:
git clone https://github.com/geissdoerfer/shepherd.git
Next, install the tools used for installing and controlling the shepherd nodes.
We'll use Ansible to remotely roll out the basic configuration to each shepherd node and shepherd-herd to orchestrate recording/emulation across all nodes.
The tools are hosted on PyPI
and require Python version >= 3.6.
You'll also need to have sshpass
installed on your machine, which is available through the package management system of all major distributions.
Install the tools using pip
:
pip3 install ansible shepherd-herd
The inventory/herd.yml
file shows an example of how to provide the host names and known IP addresses of your BeagleBones.
Adjust it to reflect your setup.
You can arbitrarily choose and assign the hostnames (sheep0, sheep1, in this example) and the ansible_user (jane in this example).
sheep:
hosts:
sheep0:
ansible_host: 192.168.1.100
sheep1:
ansible_host: 192.168.1.101
sheep2:
ansible_host: 192.168.1.102
vars:
ansible_user: jane
Now run the bootstrap.yml
Ansible playbook, which sets the hostname, creates a user and enables passwordless ssh and sudo:
ansible-playbook deploy/bootstrap.yml
Finally, use the install.yml
playbook to setup the shepherd software, optionally configuring PTP for time-synchronization:
ansible-playbook deploy/install.yml -e ptp=true
Record two minutes of data:
shepherd-herd record -l 120 -o recording.h5
The command starts the recording asynchronously and returns after all nodes have started recording. While the nodes are still recording (indicated by blinking of LED 1 and 2), prepare a directory on your local machine:
mkdir ~/shepherd_recordings
After the nodes stop blinking, you can retrieve the data to analyze it on your local machine:
shepherd-herd retrieve recording.h5 ~/shepherd_recordings
After installing some additional dependencies, you can use the provided command line utility from extra/plot.py
to visualize the recorded data:
pip install numpy matplotlib h5py click
python extra/plot.py -f ~/shepherd_recordings/sheep1/rec.h5
For a detailed description of the HDF5 based data format, refer to the corresponding documentation.
Finally, replay the previously recorded data to the attached sensor nodes, recording their power consumption:
shepherd-herd emulate -o consumption.h5 recording.h5
Try shepherd-herd --help
or check out the documentation here for a list of all options.
shepherd is an early stage project and chances are that something is not working as expected. Also, some features are not yet fully supported or documented. If you experience issues or require additional features, please get in touch via e-mail or by creating an issue on github.
The documentation is hosted on readthedocs.
shepherd development is lead at the Networked Embedded Systems Lab at TU Dresden as part of the DFG-funded project Next-IoT.
The following people have contributed to shepherd: