Archive: May 2022

Monocular vision target distance calculation.

Alongside the designing process of the Mimas Rover, computer vision experiments were carried out. A focus was made on coding an algorithm to calculate the distance of the camera from a given target, using monocular vision in OpenCV. The distance is estimated by calculating the dimension of the target and its position with respect to the centre of the frame. This algorithm will require several improvements before being actually useful for driving the Mimas Rover autonomously.

  • [Hours of work: 10h]
  • [People involved: Giorgio]

RC-VSTB, Obstacle avoidance algorithm optimization.

The team met up in room B60 at Merchiston Campus (Edinburgh), the laboratory for AC current to work on a full draft of the obstacle avoidance algorithm. We received some spare parts from other students' projects from a lecturer: we hope we might be able to give them a new life in our final project. RC-VSTB cable management has been fixed, even if there is still plenty of space for further improvements.

  • [Hours of work: 8h]
  • [People involved: Giorgio, Akshit]

enter image description here

enter image description here

Color detection.

Computer vision is one of the most exciting divisions of computer science, it consists of a series of operations made using Artificial Intelligence (AI) to process images and videos. From the perspective of engineering, its aim is to automate tasks that the human visual system can do. For this project, the OpenCV library is used. OpenCV is an Open Source library that can take advantage of multi-core processing and hardware acceleration. The first step made to approaching computer vision was to be able to detect colours, and geometrical shapes and compute simple tasks using data collected from a camera.

  • [Hours of work: 4h]
  • [People involved: Giorgio]

RC-VSTB, driving with the top down!

It's time for a second test of the RC-VSTB! This time it drove at full speed in the Orwell Terrace courtyard, steering without issues, controlled by SSH connection. All onboard systems worked as predicted and the test was successfully completed! Despite its similarity with a cabriolet car makes it look very nice, keeping all circuits exposed during outdoor tests might represent a potential hazard: so, a rearrangement of circuits and wires is compulsory. The next big improvement will be to implement an obstacle avoidance algorithm based on ultrasonic distance scanning.

  • [Hours of work: 4h]
  • [People involved: Giorgio, Akshit]

enter image description here

RC-VSTB, steering system.

Steering is essential for driving, so it was implemented successfully on the RC-VSTB together with a draft of the code to run a servo motor: the idea is to implement an appropriate sensor (Ultrasonic Distance Sensor - HC-SR04) to calculate the distance, placing it on top of a servo motor. Using only one sensor located on a rotating base, helps the RC-VSTB consume less current and run the code easier, instead of a group of sensors individually powered and connected to the Raspberry Pi 1.

  • [Hours of work: 3h]
  • [People involved: Giorgio, Akshit]

enter image description here

RC-VSTB, power up!

A self-sustaining power supply is essential for driving an autonomous vehicle: the RC-VSTB was finally disconnected from a wall socket source of energy and was provided with a battery pack. A first draft of the program that will run on the RC-VSTB has been written: no motor drivers are used in this stage.

  • [Hours of work: 18h]
  • [People involved: Giorgio, Akshit]