Titulo Estágio
JetNavigator
Local do Estágio
Coimbra
Enquadramento
Usually, a GPS is used to navigate from spot A to spot B. But as GPS doesn’t work well indoors, a different system is needed. That’s why indoor navigation or indoor wayfinding is used. This is still a recent technology and the industry is booming. The technology that’s used for indoor navigation is an indoor positioning system (IPS) or indoor location tracking. The position is calculated on the end-user’s device, hence it is called client-side positioning. Nowadays, the most cost-effective technological solution for effective indoor navigation are beacons. Beacon technology means that a space owner installs several small beacons in a building. These beacons transmit a continuous radio signal that is detected by nearby smartphones using Bluetooth. However, with technology evolving rapidly, there have been other solutions to this problem. The most recent efforts include building augmented reality navigation solutions, and localisation using Computer Vision and Deep Learning techniques.
Objetivo
The main goal of this internship is to design and implement an indoor navigation system using Computer Vision and augmented reality. In particular, the following elements must be addressed:
- Development of the autonomous robot (learn how to navigate through a building avoiding obstacles until the building has been completely mapped)
- Development of the classifier that will position the user in the building according to images
- Development of the algorithm that computes the navigation instructions
- Prototype of an application with augmented reality to render the navigation instructions
Plano de Trabalhos - Semestre 1
The main goal of this internship is to design and implement an indoor navigation system using Computer Vision and augmented reality. In particular, the following elements must be addressed:
- Development of the autonomous robot (learn how to navigate through a building avoiding obstacles until the building has been completely mapped)
- Development of the classifier that will position the user in the building according to images
- Development of the algorithm that computes the navigation instructions
- Prototype of an application with augmented reality to render the navigation instructions
Plano de Trabalhos - Semestre 2
The second semester comprises the following stages:
- Setting up the Development Environment [result: Development Environment, M6]
- Development of the autonomous robot [result: first prototype, M7]
- Development of the classifier for indoor positioning [result: DL model, M8]
- Development of the navigation algorithm and prototype app [result: app prototype, M9]
- Testing and Benchmarking [result: second prototype, M10]
- Writing the internship report [result: internship report, M10 and M11]
Condições
É fornecido computador, posto de trabalho, o robot JetRacer montado e testado, sensor de LIDAR para adicionar ao robot caso seja pretendido.
Observações
The project will be developed at Critical Software offices in Taveiro. The data used for model training (robot and classifier) will be collected on site. Pre-trained models are available that will need to be fine-tuned.
Orientador
Rui Miguel Lourenço Lopes
rui.lopes@criticalsoftware.com 📩