Driving Behaviour Analysis Software for Data-Driven Path Planning Functionalities for Automated Vehicles
2022-01-0251
03/29/2022
- Event
- Content
- Autonomous driving is currently one of the most challenging Artificial Intelligence (AI) problems as it requires combination of state-of-the-art solutions in multiple areas including computer vision, sensor fusion, control theory and software engineering. Deep learning has been pivotal to solving some of these problems, especially in computer vision. This enabled some autonomous vehicle companies started leveraging the benefits of deep learning for creating smooth, natural, human-like motion planning systems. In particular, the plethora of driving data captured from modern cars is a key enabler for training data-driven path planning systems. , Developing deep learning-powered systems relies heavily on big and high-quality data required for training of the models, in which the intrinsic statistics of the data that the model is trained on can result in different agent behavior in different scenarios. To ensure safe and comfortable limitations to the behavioral spectrum of the intelligent agent the training dataset should be balanced/conditioned on as many driving features and attributes as possible. The intelligent utilization of human driving behavioral data for improving the comfort of the passenger ride experience and the safety of deep-learning powered systems are discussed in this paper. The core idea of the proposed solution is the automatic extraction of driving features followed by the selection and balancing of key desired features and driving attributes that can then be used to update machine learning models responsible for vehicle path planning and control. For this task, a DRIVing Behavior Analysis Software (DRIVBAS) was developed with the purpose of increasing the efficiency of data analytics and machine learning activities as applied to AI-driven autonomous vehicles. The overall functionality and efficacy of the proposed solution is demonstrated with real-world experimental results.
- Citation
- SONGUR, N., and SOUFLAS, I., "Driving Behaviour Analysis Software for Data-Driven Path Planning Functionalities for Automated Vehicles," SAE Technical Paper 2022-01-0251, 2022, .