SMO - Design and implementation of the control architecture for a phenotyping robot

SMO is a machine builder providing customized machines, industrial automation and technical constructions who has invested in the development of automated systems for plant phenotyping (https://www.wiwam.be/) for agricultural purposes. Within this product line, SMO wanted to broaden their offer with a phenotyping robot capable of navigating autonomously in open fields.

Exobotic Technologies consulted SMO about the autonomous driving of their robot based on GPS, but also on the setup of a solid architecture and flexible framework.

Below the different elements of the consultancy mission.

---

System Requirements analysis

1. Analysis of received documentation

2. Clarification meeting on location

3. Processing inputs from visit

4. Definition of system requirements in common agreement with SMO

5. Presentation of system requirements document & last iteration

System Architecture definition

1. Preliminary functional architecture, prior to the start of implementation

2. Documentation of the solution at the end of the project, including:

   1. A global software architecture

   2. A data flow diagram

   3. Description of interfaces between components (ROS topics, bus interfaces, software interfaces)

   4. Functional description of each component

   5. Installation and compilation README manual

Path planning & tracking

1. Based on a list of GPS coordinates and additional user parameters, the path planner will create an optimal trajectory for navigation in the field

2. Python3-based implementation (no real-time considerations)

3. A basic robot tracking controller that generate velocity commands to follow optimally the trajectory presented in the previous page

4. Design and implementation of the controller in C++

5. Test and validation in simulation

6. Tuning and optimization of the algorithm parameters for the real robot

7. Test and validation on the real robot