Research on Autonomous Driving Scenario and Decision-making based on Deep Neural Network

The foundation of autonomous driving safety is high-precision environmental perception. As a mission-critical task, object detection has attracted considerable attention in recent years. It employs a range of sensors to gather data on factors such as target position, size, and direction of movement, thereby providing a crucial foundation for driving decisions. This study employs an enhanced Deep Convolutional Neural Network (CNN) to investigate decision-making processes in autonomous driving scenarios, thereby advancing the field of artificial intelligence. The introduction of a multi-scale feature extraction module and a self-attention mechanism serves to enhance the model's capacity for perception of complex traffic environments and recognition of key scenes. The model is capable of efficiently processing sensor data, thereby facilitating accurate scene understanding and real-time decision optimisation. The experimental analysis demonstrates that the enhanced model exhibits superior performance compared to the conventional approach across a spectrum of driving scenarios, thereby enhancing the precision and dependability of decision-making.