Research On Lane Change Fusion Decision Model For Complex Traffic

Aiming at the problem that driverless cars are unable to accurately detect the behavior of surrounding vehicles when performing lane changing in complex traffic. This paper proposes a lane changing fusion decision model to solve the problem of low effectiveness of driverless cars under complex traffic. The model improves the multi-objective trajectory prediction problem of driverless cars under complex traffic by using multi-source information fusion technology. At the same time, the input data is processed to realise the automatic learning and analysis of a large amount of data in combination with the deep learning algorithm of the artificial neural network. Statistical analysis of the model's simulation experiment outcomes was conducted, and the fusion outcomes were then contrasted to the single-sensor situation. Under the optimized lane-changing model, the driverless car is improved in both lane-changing prediction and multi-vehicle perception, which solves the problems of inaccurate multi-objective trajectory prediction and inefficient lane-changing decision-making of the driverless car during lane-changing.