Deploying AI Models at the Edge

Tevel, Spectro Cloud take flying fruit-picking robots to the next level with Kubernetes

Tevel, Spectro Cloud take flying fruit-picking robots to the next level with Kubernetes

Kubernetes is being deployed in the field — literally – by Israel-based Tevel. Tevel recently announced that they will be using Spectro Cloud’s Kubernetes platform to improve the productivity of their AI-enabled flying fruit-picking robots.

According to Tevel, their fruit-picking robots address the issue that up to 30% of the world’s fruit is not being picked on time because of difficulties with logistics and labor.

Itzik Ben Zaken, a DevOps manager at Tevel, gave a presentation at KubeCon + CloudNativeCon Europe 2023. During the presentation, he discussed the various technologies his company has worked with along its journey, including software, hardware, edge computing and Kubernetes. Pedro Oliveira, a solutions architect at Spectro Cloud, also spoke alongside him.

Tevel started collaborating with Kubernetes platform provider Spectro Cloud to address the challenges they encountered while deploying flying fruit-picking robots in diverse settings. To manage Kubernetes clusters at scale at the edge, they used the Spectro Cloud Palette Kubernetes lifecycle management platform.

The first challenge was ensuring that robots could keep running during connection disruptions. For this, Tevel needed a headless, zero-touch provisioning system with all the information provided in advance and an authenticated edge token against Palette. This system allowed devices to be registered automatically with Palette, even if not online.

Another challenge was keeping the robots up and running through any disruption. To achieve this, Tevel and Spectro Cloud implemented a distributed architecture to separate edge nodes from the control and management planes, allowing each cluster to manage itself regardless of connectivity back to headquarters.

Tevel also needed its robots to be secure in case of a burglary or theft. To solve this, Tevel and Spectro Cloud used a Trusted Platform Module (TPM) with persistent data encryption and read-only roots for extra security. This technique ensures that the disks remain encrypted if any robots get stolen.

Finally, they also needed an easy way to manage upgrades without putting too much strain on operations. They achieved this through Kairos, a container-based image that splits a disk into A/B or active and inactive partitions. If an upgrade fails, it automatically reverts to the previous configuration, so the workflow isn’t interrupted.

Tevel also stated they would use the Secure Edge-Native Architecture (SENA) framework announced at KubeCon.

SENA is a collaboration between Spectro Cloud and Intel that applies zero-trust principles for deep-root security at all layers of edge computing environments. The companies say it offers a comprehensive pathway for advanced security across the hardware, software stack, runtime and edge lifecycle.

Tevel says the collaboration between itself and Spectro Cloud has allowed them to focus on its mission of helping farmers pick every piece of fruit on time, any time of day, while also gaining real-time data.

It has also reduced their time building and managing clusters, freeing their engineers to focus on innovation around robot fruit picking, Tevel executives say.

In early 2022, Spectro Cloud made headlines for raising $40 million in a Series B funding round.

Article Topics

 |   |   |   |   |   |   | 

Comments

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Featured Edge Computing Company

Edge Ecosystem Videos

Deploying AI Models at the Edge

“Deploying

Latest News