Show simple item record

dc.contributor.advisor Moustafa, Mohamed N.
dc.contributor.author Eraqi, Hesham Mohamed
dc.date.accessioned 2021-01-19T21:49:07Z
dc.date.created 2020-12-19
dc.date.issued 2021-01-19
dc.identifier.uri http://dar.aucegypt.edu/handle/10526/5961
dc.description.abstract Steering a car through traffic is a complex task that is difficult to cast into algorithms. Therefore, researchers turn to train artificial neural networks from front-facing camera data stream along with the associated steering angles. Nevertheless, most existing solutions consider only the visual camera frames as input, thus ignoring the temporal relationship between frames. In this work, we propose a Convolution Long Short-Term Memory Recurrent Neural Network (C-LSTM), which is end-to-end trainable, to learn both visual and dynamic temporal dependencies of driving. Additionally, We introduce posing the steering angle regression problem as classification while imposing a spatial relationship between the output layer neurons. Such a method is based on learning a sinusoidal function that encodes steering angles. To train and validate our proposed model and method, we used the publicly available Comma.ai dataset. Compared with the Convolutional Neural Network (CNN)-based end-to-end direct regression method, our solution improved steering root mean square error by 35% and led to more stable steering by 87%. The end-to-end approach has demonstrated suitable vehicle control when following roads and avoiding obstacles. Conditional imitation learning (CIL) extended the end-to-end approach to allow the vehicle to take specific turns in intersections to reach a destination. Unfortunately, performance dramatically decreases when deployed to unseen environments and is inconsistent against varying weather conditions. Most importantly, the current CIL fails to avoid static road blockages. In this work, we propose a solution to those deficiencies. First, we fuse the laser scanner with the regular camera stream, at the features level, to overcome the generalization and consistency challenges. Second, we introduce a new efficient Occupancy Grid Mapping (OGM) method along with new algorithms for road blockages avoidance and global route planning. Consequently, our proposed method dynamically detects partial and full road blockages and guides the controlled vehicle to another route to reach the destination. Following the original CIL work, we demonstrated the effectiveness of our proposal on CARLA simulator urban driving benchmark. Our experiments showed that our model improved CIL consistency against weather conditions by four times and autonomous driving success rate generalization by 52%. Furthermore, our global route planner improved the driving success rate by 37%. Our proposed road blockages avoidance algorithm improved the driving success rate by 27%. Finally, the average kilometers traveled before a collision with a static object increased by 1.5 times. en_US
dc.description.sponsorship Yousef Jameel '68 PhD in Applied Sciences and Engineering Fellowship en_US
dc.format.extent 122 p. en_US
dc.format.medium theses en_US
dc.language.iso en en_US
dc.rights Author retains all rights with regard to copyright. en
dc.subject Autonomous Driving en_US
dc.subject Self-driving Cars en_US
dc.subject Artificial Intelligence en_US
dc.subject Deep Learning en_US
dc.subject Mapping en_US
dc.subject Planning en_US
dc.subject Machine Learning Regression en_US
dc.subject End-to-end Learning en_US
dc.subject Imitation Learning en_US
dc.subject Behavior Cloning en_US
dc.subject Probabilistic Robotics en_US
dc.subject.classification Book en_US
dc.title End-to-end autonomous driving: a dynamic conditional imitation learning approach considering temporal dependencies en_US
dc.type Text en_US
dc.subject.discipline Computer Engineering en_US
dc.rights.access This item is restricted for 6 months from the date issued en_US
dc.contributor.department American University in Cairo. Dept. of Computer Science and Engineering en_US
dc.embargo.lift 2021-07-18T21:49:07Z
rft.btitle Parts are to be published and other parts are published in: IEEE 23rd International Conference on Intelligent Transportation Systems (IEEE ITSC), September, 2020 - Workshop on Autonomous Driving in the 34th Conference on Neural Information Processing Systems (NeurIPS), 2020 - Machine Learning for Intelligent Transportation Systems Workshop in the 31st Conference on Neural Information Processing Systems (NIPS), Long Beach, CA, USA, 2018 - Journal of Advanced Transportation, Machine Learning in Transportation (MLT) Issue, 2019 - IEEE Transactions on Intelligent Transportation Systems - IEEE 20th International Conference on Intelligent Transportation Systems (IEEE ITSC), Yokohama, Japan, October, 2017 - Communications of the ACM, 2020 en_US


Files in this item

Icon

This item appears in the following Collection(s)

  • Student Scholarship [703]
    This collection includes presentations, research findings, publications, and projects authored by AUC students

Show simple item record