Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Efficient End-to-end Autonomous Driving
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 c...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
AUC Knowledge Fountain
2020
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613419092312064 |
|---|---|
| access_status_str | Open Access |
| author | Eraqi, Hesham |
| author_browse | Eraqi, Hesham |
| author_facet | Eraqi, Hesham |
| author_sort | Eraqi, Hesham |
| collection | Thesis |
| description | 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 \cite{bojarski2016end}, 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 \cite{codevilla2018end} 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. |
| format | Thesis |
| id | oai:fount.aucegypt.edu:etds-2543 |
| institution | American University in Cairo (Egypt) |
| last_indexed | 2026-06-10T12:35:50.652Z |
| license_str | Not specified — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from AUC Knowledge Fountain — bepress |
| publishDate | 2020 |
| publishDateRange | 2020 |
| publishDateSort | 2020 |
| publisher | AUC Knowledge Fountain |
| publisherStr | AUC Knowledge Fountain |
| record_format | dspace |
| source_str | AUC Knowledge Fountain — bepress |
| spelling | oai:fount.aucegypt.edu:etds-2543 Efficient End-to-end Autonomous Driving Eraqi, Hesham 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 \cite{bojarski2016end}, 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 \cite{codevilla2018end} 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. 2020-12-19T08:00:00Z dissertation application/pdf https://fount.aucegypt.edu/etds/1534 https://fount.aucegypt.edu/context/etds/article/2543/viewcontent/hesham_mohamed_eraqi_thesis.pdf Theses and Dissertations AUC Knowledge Fountain Autonomous Driving Sensor Fusion Occupancy Grid Mapping Route Planning Work Zones Detection Deep Learning End-to-end Learning Autonomous Driving Conditional Imitation Learning Temporal Dependencies Regression and Classification Navigation, Guidance, Control, and Dynamics Other Computer Engineering Robotics |
| spellingShingle | Autonomous Driving Sensor Fusion Occupancy Grid Mapping Route Planning Work Zones Detection Deep Learning End-to-end Learning Autonomous Driving Conditional Imitation Learning Temporal Dependencies Regression and Classification Navigation, Guidance, Control, and Dynamics Other Computer Engineering Robotics Eraqi, Hesham Efficient End-to-end Autonomous Driving |
| title | Efficient End-to-end Autonomous Driving |
| title_full | Efficient End-to-end Autonomous Driving |
| title_fullStr | Efficient End-to-end Autonomous Driving |
| title_full_unstemmed | Efficient End-to-end Autonomous Driving |
| title_short | Efficient End-to-end Autonomous Driving |
| title_sort | efficient end to end autonomous driving |
| topic | Autonomous Driving Sensor Fusion Occupancy Grid Mapping Route Planning Work Zones Detection Deep Learning End-to-end Learning Autonomous Driving Conditional Imitation Learning Temporal Dependencies Regression and Classification Navigation, Guidance, Control, and Dynamics Other Computer Engineering Robotics |
| url | https://fount.aucegypt.edu/etds/1534 https://fount.aucegypt.edu/context/etds/article/2543/viewcontent/hesham_mohamed_eraqi_thesis.pdf |
| work_keys_str_mv | AT eraqihesham efficientendtoendautonomousdriving |