Incremental End-to-End Learning for Lateral Control in Autonomous Driving

Abstract

The use of high-quality data is required to complete the job of lateral control utilizing Behavioral Cloning (BC) through an End-to-End (E2E) learning system. The majority of E2E learning systems gather this high-quality data all at once before beginning the training phase (i.e., the training process does not start until the end of the data collection process). The demand for high-quality data necessitates a large amount of human effort and substantial time and money spent waiting for data collection to be completed. As a result, it is critical to find a viable option to reduce both the time and cost of data collecting while also maintaining the performance of a trained vehicle controller. This paper offers a novel behavioral cloning approach for lateral vehicle control to address the aforementioned problems. The proposed technique begins by collecting the least amount of human driving data possible. The data from human drivers are utilized for training a convolutional neural network for lateral control. The trained neural network is subsequently deployed to the vehicle's automated driving controller, replacing a human driver. At this point, a human driver is out of the loop, and an automated driving controller, trained by the initial data from a human driver, drives the vehicle to collect further training data. The driving data obtained are sent into a convolutional neural network training module, then the newly trained neural network is deployed to the automated driving controller that will drive the vehicle further. The data collection alternates neural network training processes using the collected data until the neural network learns to correctly associate an image input with a steering angle. The proposed incremental approach was extensively tested in simulated environments, and the results are promising, only 3.81% (1,061 out of 27,884) of the total data came from a human driver. The incrementally trained neural networks using data collected by automated controllers were able to drive the vehicle in two different tracks successfully. The AI chauffeur was able to drive the vehicle on Track B for more than 70% of the track even though it has not seen the track before.

OPEMI: Online Performance Evaluation Metrics Index for Deep Learning-Based Autonomous Vehicles

“…End-to-end imitation learning was used to test the proposed metrics using the OSCAR open-source platform, which is a full-stack open-source robotic car architecture designed to enhance robotic research and teaching in AVs [5].…”

“…We designed two different tracks in the Gazebo world format [5] [15], one for training and the other for testing. Track-1 was used to collect data for training the DNNs, as shown in Fig.…”

“…Bimi-Net was proposed in [5], it has five convolutional layers and five fully connected layers with no no maximum pooling layers BiMiwL Similar to Bimi-Net but we changed the input image shape to a square and added an LSTM layer the limitations of both offline and online metrics are thereby addressed. The first set was created to highlight the shortcomings of the traditional measures and offer superior alternatives.…”

“…BiMi-Net and BiMiwL are our models. BiMi-Net was proposed in our previous work [5]. BiMiwL has similar architecture to BiMi-Net with one added LSTM layer and the input image shape is set to square.…”

“…The majority of comparative DNN model studies for AVs use prediction accuracy metrics (offline) [1]- [4]. Driving performance metrics (online) have often been used to measure how far an AV drives without human intervention [5]- [7]. Fig.…”