Published Paper Details:

The development of a road-following and collision-avoidance system uses the Jetson Nano-powered JetBot. JetBot is an open-source robot platform designed for learning and experimentation in robotics and AI, equipped with various sensors and cameras to facilitate autonomous navigation. This project focuses on developing a robust system for autonomous vehicles that can effectively follow roads and avoid collisions. The system integrates road detection with real-time obstacle avoidance using advanced machine learning and sensor technologies such as computer vision, LiDAR, and ultrasonic sensors. A convolutional neural network (CNN) identifies lane markings and road boundaries for accurate road following, while reinforcement learning and path planning algorithms ensure proactive collision avoidance. Simulations and real-world tests demonstrate significant improvements in navigation reliability across diverse environments, advancing the safety and efficiency of autonomous driving technologies. This project aims to develop an integrated system for autonomous vehicles focusing on road following and collision avoidance. Key components include: 1. Data Collection and Preprocessing: Gather diverse datasets of road images, LiDAR and ultrasonic data. 2. Road Following System: Train a CNN for lane and boundary detection. 3. Collision Avoidance System: Implement reinforcement learning and path planning for obstacle avoidance. 4. Sensor Fusion: Integrate data from cameras, LiDAR and ultrasonic for situational awareness. 5. Simulation and Testing: Validate system performance through simulations and real-world testing. 6. Optimization: Refine algorithms and parameters based on testing results. Finally, the outcome of this project is to Integrate the Road Following and Collision Avoidance models to allow Jetbot to autonomously work with an average speed while simultaneously avoiding collisions with obstacles. Furthermore, provide smooth system-to-system communication for effective operation and real-time decision-making. Use thorough testing procedures, including worst-case situations, to find any possible flaws. Use cutting-edge machine learning strategies to improve system performance over time. Create a cooperative atmosphere that promotes creativity and information exchange. Establishing a feedback loop with end users will also assist in quickly identifying and resolving any real-world issues. The Jetbot will handle the needs of dynamic and complicated settings by concentrating on these factors, which will lead to increased autonomy and safety.