Driving Intention Inference Based on a Deep Neural Network with Dropout Regularization from Adhesion Coefficients in Active Collision Avoidance Control Systems
Driving intention, which can assist drivers to avoid dangerous emergence for the advanced driver assistant systems (ADAS), can be hardly described accurately for complex traffic environments. At present, driving intention can be mainly obtained by deep neural networks with neuromuscular dynamics and...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-07-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/15/2284 |
| Summary: | Driving intention, which can assist drivers to avoid dangerous emergence for the advanced driver assistant systems (ADAS), can be hardly described accurately for complex traffic environments. At present, driving intention can be mainly obtained by deep neural networks with neuromuscular dynamics and electromyography (EMG) signals of drivers. This method needs numerous drivers’ signals and neural networks with a complex structure. This paper proposes a driving intention direct inference method, namely direct inference from the road surface condition. A driving intention safety distance model based on a deep neural network with dropout regularization was built in an active collision avoidance control system of electric vehicles. Driving intention can be inferred by a deep neural network with dropout regularization from adhesion coefficients between the tire and road. Simulations using rapid control prototyping (RCP) and a hardware-in-the-loop (HIL) simulator were performed to demonstrate the effectiveness of the proposed driving intention safety distance model based on a deep neural network with dropout regularization. The proposed driving intention safety distance model can guarantee the safe driving of electric vehicles. |
|---|---|
| ISSN: | 2079-9292 |