Using autonomous farm equipment is widely viewed as a partial solution to the labour shortages plaguing the agri-food sector. However, one of the barriers to the adoption of such technology is its tendency to rely solely on GPS-driven navigation controllers, which are adequate to the task only in environments without unplanned obstacles. In effect, this means an autonomous tractor is not truly independent, as it requires the constant presence and vigilance of a trained human operator, defeating the purpose of the so-called autonomy.

Edmonton-based Mojow’s EYEBOX™ technology is a small, rugged, economical sensor suite outfitted with multiple cameras, a GPS, and a powerful computer to process data in real time. The system functions by collecting images automatically and passing them through a variety of algorithms that classifies each pixel to create (or update) a digital representation of the entire farming entity (field boundaries, roads, field entrances, or anything else of interest). This electronic “twin” of a farm’s physical attributes serves as the foundation (or primary input) of the autonomous navigation controller—the “brain,” if you will, that makes EYEBOX-guided machinery truly autonomous. The continuous intake of real time image data from the peripherals of the tractor assures a high level of relative position accuracy between the vehicle and any physical object encountered within its working environment.

As part of this project, the team intends to develop the following autonomous technology key enablers:

  • Real-time detection of, and response to, all relevant external and internal field boundaries, requiring the development of deep-learning models and software algorithms.
  • Real-time detection of, and response to, all types of field entrance, specifically
  • Small-width entrances,
  • Larger-width entrances, and
  • Single-field segment to segment.

This will permit a tractor, without any human interaction, to identify, predict, and choose an appropriate route from roadway to field, field to roadway, or from one segment to another within a single field. The EYEBOX will save time by appropriately planning and executing the most efficient entrance and exit locations.

  • Real-time detection of, and response to, all roadway types, specifically dirt and gravel roads and double track trails. Successfully programming autonomous field-to-field transition is a necessary system capability if EYEBOX is to achieve its goal of freeing farmers from their tractors.

EYEBOX is designed to operate ISO 11783-certified farm implements such as air seeders, planters, sprayers, and fertilizer spreaders. However, Mojow intends to start with land rolling and heavy harrowing, thereby proving its concept before adding tools that apply product.

Designed for flexibility, this technology will be able both to convert conventional tractors into autonomous vehicles and be integrated into OEM machinery to enhance functionality.

When fully operational, the EYEBOX platform will lower a farm’s production costs while increasing output and reducing reliance on manual labour.

CAAIN Contribution
$631,981

Total Project Value
$2,083,056

Project Contact
Owen Kinch
Mojow Autonomous Solutions
owen@mojow.ai