What is Edge computing?
Edge computing involves the processing and storage of data closer tothe sources of the data, which takes place at the ‘edge’ of the network.
Benefits of computing at the edge can range fromsaving resources, increased security, processing speed, more refined privacyand data management measures.
What is Zigbee?
Zigbee is a wireless communication protocol that operates on the2.4GHz radio band and is AES128 Encrypted. Zigbee enables the creation of localarea/proximity networks with smaller, low-power and low data rate devices.
Devices that connect to the same hub in an ecosystem will be able tocreate a mesh network by themselves to have a stronger network and range in theecosystem.
Devices can also communicate by hopping between each node (5m-15mestimated) per hop to relay messages back and forth between the devices if theyare not within direct communication range of the hub.
Gravio is using the Zigbee 3.0 stack:
How do I pair a Zigbee sensor?
To pair a Zigbee device on Gravio you need to put the Gravio HubKit into pairing mode, bring the devices sufficiently close (a few cm/inches) and trigger the stimulus button a few times, shortly after the device should appear in the devices list.
Note: you may have to repeat this process multiple times because the sensors pair once enough data has been exchanged and validated. And if the sensor has been paired before with a different HubKit, you will need to reset the pairing by long-pressing the stimulus button for 5+ seconds until the light is set for a longer time.
The binding process creates a logical association between two Zigbee nodes, aka a pairing link. Nodes within a Zigbee network may only exchange data messages directly with other devices on the network after the binding process has completed successfully.
The binding process consists of 4 stages:
A Zigbee device can perform discovery in an attempt to find other suitable devices for pairing. Discovery can be attempted repeatedly on all three channels for a fixed duration or until a sufficient number of responses have been received. Service discovery is only available to devices that are not currently in power-saving mode. During discovery, a number of pieces of information are exchanged between both devices. This information is passed to the application, which can then make a decision whether it should respond.
Once a device has determined, through discovery, that there is another device within communication range offering compatible services, it can set up a pairing link in order to begin communication.
The Zigbee standard supported by Gravio requires a “push button” stimulus on the target device to start the pairing process. After successful discovery of other devices within communication range, the pairing originator ranks the devices based on their class descriptor(s) to find the best pairing candidate. It attempts pairing with the highest ranked device, if not successful, it attempts pairing with the second highest ranked device, and so on. This pairing is temporary and used for configuration and interactive validation. Note that security is enabled for the temporary pairing to ensure an encrypted communication channel for the configuration and validation phases.
How do I install Gravio on my computer?
Gravio requires mainly 2 pieces of software:
● The Gravio HubKit (which is the server)
Gravio HubKit and Gravio Studio are usually not installed on the same machine, but this is still possible.
Also, you can have multiple HubKit software on multiple different devices including Mac, Windows and Ubuntu Linux (Including Raspberry Pi). You can download the appropriate version of your server software from the Gravio Website.
Once you have installed both Gravio Studio (for management) and Gravio HubKit (as edge server) on the same network, you can start by creating a login account on Gravio Studio. Just launch your Gravio Studio and follow the instructions.
What hardware and software does Gravio operate on?
The main two Gravio pieces of software that you require, i.e. theGravio Studio to configure your setup including your actions, and the servercalled Gravio HubKit to run the operations at the edge can be installed on thefollowing hardware and software respectively:
● Gravio Studio runs on Macos andWindows. You can download them from the respective App Stores. We alsorecommend the latest updates of the respective operating systems, although theminimum requirements are visible in the App Store metadata.
The Gravio HubKit (i.e. the Edge Server) runs onmacOS, Windows and Linux (Ubuntu, 64 bit, as well as the Raspberry Pi 4 withUbuntu). Please refer to the respective minimum requirements at the pointof download on our website.
What is the difference between a Sensor and an Actuator?
A Sensor is a device that measures and provides an output value that is sent to the Hub/PC. This information will be reflected in Gravio Studio which can be used to Trigger events.
An Actuator is a piece of hardware that has an input that can be triggered or activated by Gravio. Examples of actuators are Gravio Lights, the Gravio LED Matrix, or any device that can be activated by an API that Gravio supports.
When we say actuator, do we mean devices likethe Aqara button? Or do we mean Actions within Gravio Studio?
How does Computer vision work?
Computer vision includes the ability for computers to acquire, process and analyze digital images or videos, from which they can gain an understanding and provide a structured representation of that information.
Which Gravio package do you recommend for which application?
If you would like to familiarise yourself with Gravio, it’s architecture and user interface, we suggest something that does not require any hardware. The free version supports some basic computer vision models that can be activated by a connected network (ONVIF) or USB camera. It is not fit for any production environment as the authentication systems are not enabled for this version.
If you like to try 3-4 basic hardware sensors to see how they behave, we suggest you go for the basic version, which you can sign up for in the Gravio Studio app downloaded from the Windows or Mac App Store.
The standard package is recommended for production environments. It comes with 20 sensors of your choice. But the computer vision models are still limited to what we provide for you.
The Enterprise package is the full package and includes a back-end for you, should you require one. The Enterprise package is also the only package that allows you to create your own computer vision models for your deployment.
Please Note: the devices are rental devices and we are subject to import/export regulations in order to be able to ship those devices. If you are unsure if the sensors are available in your country, don’t hesitate to send us an email to firstname.lastname@example.org. Similarly to the free package, the basic package is not recommended for production environments.
What are other people building with Gravio?
What do Low-code or No-code mean?
The term low-code platform (also low-code development platform) describes a development environment for software that enables development with visual application design tools and other graphical modelling techniques instead of using text-based programming languages.
This can significantly reduce the development and deployment time for software. Because of these potential benefits, low-code platforms can significantly reduce the costs of project planning, staff training and actual development.
Usually, low-code development is carried out in an integrated development environment. Unlike in conventional programming environments, prefabricated standard software modules are often used instead of individual programme code. Metadata such as variables and other parameters are then used to link these in such a way that the desired programme behaviour is created.
High customer interest in low-code development platforms has led to a dynamic market landscape.
This growth is fuelled by the increasing appreciation of low-code platforms as tools for developing new technologies and existing technologies in large enterprises.
More specifically, a distinction can be made between low-code automation, low-code app development and low-code prototyping.
● Low-code automation is characterised by the fact that existing SaaS can be connected with APIs. Users can thus build on existing systems and connect familiar app front-end applications with each other.
● Low-code app development, on the other hand, refers to the creation of a front-end and the associated back-end logic. However, this could be replicated by iPaaS to lower the complexity of low-code app development.
● In low-code prototyping, only central functions and GUIs are realized; as a rule, no market-ready application is created.
How do I create my own Computer Vision models?
You can follow thesteps in our documentation to create your own computer vision models. Butplease bear in mind that bespoke computer vision models are only available inthe Enterprise edition of Gravio. For support or questions, don’t hesitate to join our Slackchannel and ask your question there.
What makes Gravio unique compared to other platforms?
Gravio distincts itself from other IoT platform in the followingpoints:
● Edge-first: Gravio was built edge-first.That means all data is processed and actuators are triggered on-site at theedge. We believe that the brain of a smart space should be within that space inthe future. This has certain security, data ownership, privacy and efficiency advantages.Having said that, Gravio can also integrate with cloud platforms. Storing data,showing information on a dashboard or triggering SMS/messaging notificationsare examples for functionalities that are still best conducted at the cloudlevel.
● No internet connection required i.e. works offline: As a result of being a cloud-first platform, Gravio can workcompletely offline without any internet connection whatsoever. This can beuseful in various contexts such as on the move, in secure environments or atsea.
● Hardware and Software Sensors included:To get started with IoT quickly, Gravio comes with a set of hardware andsoftware sensors and actuatorsavailable. This makes it easy for any customer to get up to speed from thebeginning easily. Hardware sensors are physical devices connected via wirelesscommunications. Software sensors are AI/ML models that process camera feeds todetect triggerable events. Most traditional platforms tend to support justeither software sensors or hardware sensors. Gravio however brings them alltogether in one platform.
● Supported OSes: Gravio’s server runs onWindows, macOS and Linux. Most other platforms are limited to less operatingsystems. For a Gravio deployed infrastructure, it does not matter on which OSthe various edge nodes are operating. They can all function side-by-side.
● Interoperability: At Gravio we have thephilosophy to support standard protocols available. We are not in the positionto push our own standard through, which is why we decided to support availablestandards from the outset and build on those. These can be HTTP APIs, MQTT,ONVIF, Zigbee, BLE, TensorFlow etc.