Autonomous Driving Analysis, sumarry And Conclusion

Materials and Setup

• Jetson Nano Developer Kit
• USB Gamepad
• MATLAB® 2020B (compatible with CUDA 10.2) [1]
• Wi-Fi connection between Jetson Nano and host PC

Experiment Design and Requirements

• MATLAB® was used to connect to the Jetson Nano via SSH.
• The Gamepad receiver had to be physically connected to the Jetson Nano.
• MATLAB® controlled the Gamepad indirectly using SSH commands to access libraries on the Jetson.
• Data such as images and steering values were collected and transferred via Wi-Fi.

[[File:matlab_ssh_code

.jpg|frame|Fig. 3: Ssh code snipped to connect using MATLAB and JetRacer]]

Experiment Procedure

Step 1: System Connection and Gamepad Control

Since MATLAB® could not directly access the Gamepad receiver, an SSH command method was developed to control the Jetson Nano libraries from the background without delay or signal loss.

Step 2: Data Collection

• Data was collected using the integrated camera at 256×256 pixel resolution.
• Real-time steering commands were logged and synchronized with images.
• To reduce image distortion, camera calibration was performed using a calibration256by256.mat parameter.

Fig. 5: Camera setup for image without calibration

[[File:

.png|frame|Fig. 6: Camera setup for image with calibration]]

Step 3: Neural Network Training

• Initially, a pretrained network was used with RGB images, but results were poor. with the following parameters:

• Regression methods were tried but also showed poor performance. with the following parameters:

|Fig. 9: regression_50_epoch2]]

|frame|Fig. 10: regression_50_epoch]]

• The project switched to **classification** using **discrete steering values**. with the following parameters:

• MATLAB® Apps (GUI) were used to manually annotate images by clicking the correct path.

Step 4: System Optimization

• Lane tracking algorithms were integrated to automatically label lane positions.
• Additional features like lane angles were extracted to improve model inputs.
• Grayscale image recordings were used to achieve better generalization and faster processing.
• Still using **classification**. with the following parameters:

[[File:

.png|frame|Fig. 13: Camera setup for image recording during driving]]

Experiment Analysis

Training Results

• Models trained on discrete steering values showed significantly better results than regression-based approaches.
• Grayscale recordings increased model stability.
• Calibrated images helped reduce the steering error caused by lens distortion.

System Performance

• A maximum speed limit (~1 m/s) was successfully enforced via software.
• The JetRacer could autonomously drive several laps counterclockwise in the right lane.
• However, simultaneous recording, steering, and saving caused performance bottlenecks due to the Jetson Nano’s limited resources.

Limitations

• Hardware constraints limited the maximum processing speed.
• Initial data quality was poor due to randomized manual steering input.
• Better results were obtained after switching to discrete, consistent control signals.

Summary and Outlook

This project demonstrated that it is possible to optimize autonomous navigation on the JetRacer using a combination of classical control algorithms and neural networks. Despite hardware limitations, stable autonomous driving behavior was achieved.

Future improvements could include:

• Expanding the dataset with more diverse environmental conditions.
• Flash working Model directly on jetson nano using gpu coder
• Use accurate gamepad and use more complex training method.
• Improving automated labeling and refining the PD controller parameters for faster driving without loss of robustness.