Accès libre

1/10th scale autonomous vehicle based on convolutional neural network

Avishkar Seth

Avishkar Seth

Macquarie UniversitySydney, Australia
Recherchez cet auteur sur
Sciendo|Google Scholar
Seth, Avishkar
, 
Alice James

Alice James

Macquarie UniversitySydney, Australia
Recherchez cet auteur sur
Sciendo|Google Scholar
James, Alice
 et 
Subhas C. Mukhopadhyay

Subhas C. Mukhopadhyay

Macquarie UniversitySydney, Australia
Recherchez cet auteur sur
Sciendo|Google Scholar
Mukhopadhyay, Subhas C.
  
25 août 2020
1/10th scale autonomous vehicle based on convolutional neural network's Cover Image
À propos de cet article
Article précédent
Article suivant
Citez
Télécharger la couverture
Catégorie d'article: Research-Article
Publié en ligne: 25 août 2020
Pages: 1 - 17
Reçu: 26 juil. 2020
DOI: https://doi.org/10.21307/ijssis-2020-021
Mots clés
© 2020 Avishkar Seth published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1:

Picture of the individual vehicle hardware components used in the system.
Picture of the individual vehicle hardware components used in the system.

Figure 2:

Complete system circuit diagram.
Complete system circuit diagram.

Figure 3:

CAD design for the laser cut base plate implemented on Autodesk Fusion 360 (left) and the base plate placed on the vehicle chassis (right).
CAD design for the laser cut base plate implemented on Autodesk Fusion 360 (left) and the base plate placed on the vehicle chassis (right).

Figure 4:

CAD design for the tower type case implemented on Autodesk Fusion 360 (left) and the physical PLA material 3D printed case (right).
CAD design for the tower type case implemented on Autodesk Fusion 360 (left) and the physical PLA material 3D printed case (right).

Figure 5:

The overview of the software message queue of the system (vehicle state).
The overview of the software message queue of the system (vehicle state).

Figure 6:

Windows host PC installation steps.
Windows host PC installation steps.

Figure 7:

Software configuration steps for the Raspberry Pi 4 setup.
Software configuration steps for the Raspberry Pi 4 setup.

Figure 8:

Screenshot of the camera code configuration and sample 160 × 120 pixels image from the vehicle on board camera.
Screenshot of the camera code configuration and sample 160 × 120 pixels image from the vehicle on board camera.

Figure 9:

Screen capture of the I2C detected on the Rpi terminal.
Screen capture of the I2C detected on the Rpi terminal.

Figure 10:

The Localhost web interface (screen capture) to control the vehicle.
The Localhost web interface (screen capture) to control the vehicle.

Figure 11:

The complete assembly of the vehicle hardware system.
The complete assembly of the vehicle hardware system.

Figure 12:

The code snippet program of the motor calibration.
The code snippet program of the motor calibration.

Figure 13:

The steering servo and PWM working principle.
The steering servo and PWM working principle.

Figure 14:

Table generated for steering Angle and PWM equivalent values.
Table generated for steering Angle and PWM equivalent values.

Figure 15:

The custom-built indoor track design and its implementation.
The custom-built indoor track design and its implementation.

Figure 16:

A sample training dataset from the track training.
A sample training dataset from the track training.

Figure 17:

Training data model loss graph.
Training data model loss graph.

Figure 18:

The car algorithm flowchart.
The car algorithm flowchart.

Figure 19:

Training and autopilot steering angle histogram graph plot analysis.
Training and autopilot steering angle histogram graph plot analysis.

No. Convolution filters Strides FC layers Parameters Loss
1 12 × 3 × 3, 18 × 3 × 3, 24 × 3 × 3, 36 × 3 × 3 3,2,2,1 900, 246, 32 1,100k 0.098543
Langue:
Anglais
Périodicité:
1 fois par an
Sujets de la revue:
Ingénierie, Présentations et aperçus, Ingénierie, autres
RSS Feed de la revue