Derek Rayside
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Derek Rayside has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Mohammad Al-Sharman
Department of Electrical and Computer Engineering and the Waterloo Autonomous Vehicle Team in the SAE AutoDrive Challenge (WATonomous) (Watonomous.ca), University of Waterloo, Waterloo, Canada
2 shared publications - William Melek
Department of Mechanical and Mechatronics Engineering, University of Waterloo, Waterloo, ON, Canada
2 shared publications - Alex Zhuang1 shared publication
- Alikasim Budhwani
Mechanical and Mechatronics Engineering Department and the Waterloo Autonomous Vehicle Team in the SAE AutoDrive Challenge (Watonomous.ca) (WATonomous), University of Waterloo, Waterloo, ON, Canada
1 shared publication - Carl Thomas Haas
Univ. of Waterloo
1 shared publication - Chris Rennick
National Physical Laboratory, Teddington, UK
1 shared publication
Latest Publications
- Open Access
RALACs: Action Recognition in Autonomous Vehicles using Interaction Encoding and Optical Flow
When applied to autonomous vehicle (AV) settings, action recognition can enhance an environment model's situational awareness. This is especially prevalent in scenarios where traditional geometric descriptions and heuristics in AVs are insufficient. However, action recognition has traditionally been studied for humans, and its limited adaptability to noisy, un-clipped, un-pampered, raw RGB data has limited its application in other fields. To push for the advancement and adoption of action recognition into AVs, this work proposes a novel two-stage action recognition system, termed RALACs. RALACs formulates the problem of action recognition for road scenes, and bridges the gap between it and the established field of human action recognition. This work shows how attention layers can be useful for encoding the relations across agents, and stresses how such a scheme can be class-agnostic. Furthermore, to address the dynamic nature of agents on the road, RALACs constructs a novel approach to adapting Region of Interest (ROI) Alignment to agent tracks for downstream action classification. Finally, our scheme also considers the problem of active agent detection, and utilizes a novel application of fusing optical flow maps to discern relevant agents in a road scene. We show that our proposed scheme can outperform the baseline on the ICCV2021 Road Challenge dataset and by deploying it on a real vehicle platform, we provide preliminary insight to the usefulness of action recognition in decision making.
Open AccessRALACs: Action Recognition in Autonomous Vehicles using Interaction Encoding and Optical Flow
When applied to autonomous vehicle (AV) settings, action recognition can enhance an environment model's situational awareness. This is especially prevalent in scenarios where traditional geometric descriptions and heuristics in AVs are insufficient. However, action recognition has traditionally been studied for humans, and its limited adaptability to noisy, un-clipped, un-pampered, raw RGB data has limited its application in other fields. To push for the advancement and adoption of action recognition into AVs, this work proposes a novel two-stage action recognition system, termed RALACs. RALACs formulates the problem of action recognition for road scenes, and bridges the gap between it and the established field of human action recognition. This work shows how attention layers can be useful for encoding the relations across agents, and stresses how such a scheme can be class-agnostic. Furthermore, to address the dynamic nature of agents on the road, RALACs constructs a novel approach to adapting Region of Interest (ROI) Alignment to agent tracks for downstream action classification. Finally, our scheme also considers the problem of active agent detection, and utilizes a novel application of fusing optical flow maps to discern relevant agents in a road scene. We show that our proposed scheme can outperform the baseline on the ICCV2021 Road Challenge dataset and by deploying it on a real vehicle platform, we provide preliminary insight to the usefulness of action recognition in decision making.
- Open Access
Sim-to-Real Domain Adaptation for Lane Detection and Classification in Autonomous Driving
While supervised detection and classification frameworks in autonomous driving require large labelled datasets to converge, Unsupervised Domain Adaptation (UDA) approaches, facilitated by synthetic data generated from photo-real simulated environments, are considered low-cost and less time-consuming solutions. In this paper, we propose UDA schemes using adversarial discriminative and generative methods for lane detection and classification applications in autonomous driving. We also present Simulanes dataset generator to create a synthetic dataset that is naturalistic utilizing CARLA's vast traffic scenarios and weather conditions. The proposed UDA frameworks take the synthesized dataset with labels as the source domain, whereas the target domain is the unlabelled real-world data. Using adversarial generative and feature discriminators, the learnt models are tuned to predict the lane location and class in the target domain. The proposed techniques are evaluated using both real-world and our synthetic datasets. The results manifest that the proposed methods have shown superiority over other baseline schemes in terms of detection and classification accuracy and consistency. The ablation study reveals that the size of the simulation dataset plays important roles in the classification performance of the proposed methods. Our UDA frameworks are available at https://github.com/anita-hu/sim2real-lane-detection and our dataset generator is released at https://github.com/anita-hu/simulanes
Open AccessSim-to-Real Domain Adaptation for Lane Detection and Classification in Autonomous Driving
While supervised detection and classification frameworks in autonomous driving require large labelled datasets to converge, Unsupervised Domain Adaptation (UDA) approaches, facilitated by synthetic data generated from photo-real simulated environments, are considered low-cost and less time-consuming solutions. In this paper, we propose UDA schemes using adversarial discriminative and generative methods for lane detection and classification applications in autonomous driving. We also present Simulanes dataset generator to create a synthetic dataset that is naturalistic utilizing CARLA's vast traffic scenarios and weather conditions. The proposed UDA frameworks take the synthesized dataset with labels as the source domain, whereas the target domain is the unlabelled real-world data. Using adversarial generative and feature discriminators, the learnt models are tuned to predict the lane location and class in the target domain. The proposed techniques are evaluated using both real-world and our synthetic datasets. The results manifest that the proposed methods have shown superiority over other baseline schemes in terms of detection and classification accuracy and consistency. The ablation study reveals that the size of the simulation dataset plays important roles in the classification performance of the proposed methods. Our UDA frameworks are available at https://github.com/anita-hu/sim2real-lane-detection and our dataset generator is released at https://github.com/anita-hu/simulanes
- Open Access
- 6 November 2019
- Published by Queen's University Library in Proceedings of the Canadian Engineering Education Association (CEEA)
SOFTWARE ENGINEERING DAYS: USING A VIDEO GAME PLATFORM TO TEACH COLLABORATIVE SOFTWARE DEVELOPMENT
This paper describes the implementation and assessment of a multi-day engineering design activity for Software Engineering students. This activity required students to work in teams of 16 to develop and implement four sub-systems of a digital spaceship. The Unity video game engine with a custom "spaceship sandbox" were used to drive student intrinsic motivation for the task, and to limit the complexity of the spaceship implementation. Student feedback of the activity was captured through a survey given immediately following the second day of the activity, and through a focus group conducted with four students later in the term. The feedback on the activity was largely positive, with teamwork, collaboration, and the software to accomplish it (Git) as major learning outcomes identified by students. Several improvements are planned for the fall 2019 version of the activity based on both instructor observation and student feedback.
- 6 November 2019
- Published by Queen's University Library in Proceedings of the Canadian Engineering Education Association (CEEA)
Open AccessSOFTWARE ENGINEERING DAYS: USING A VIDEO GAME PLATFORM TO TEACH COLLABORATIVE SOFTWARE DEVELOPMENT
This paper describes the implementation and assessment of a multi-day engineering design activity for Software Engineering students. This activity required students to work in teams of 16 to develop and implement four sub-systems of a digital spaceship. The Unity video game engine with a custom "spaceship sandbox" were used to drive student intrinsic motivation for the task, and to limit the complexity of the spaceship implementation. Student feedback of the activity was captured through a survey given immediately following the second day of the activity, and through a focus group conducted with four students later in the term. The feedback on the activity was largely positive, with teamwork, collaboration, and the software to accomplish it (Git) as major learning outcomes identified by students. Several improvements are planned for the fall 2019 version of the activity based on both instructor observation and student feedback.
- 1 September 2016
- Published by American Society of Civil Engineers (ASCE) in Journal of Computing in Civil Engineering
- Vol. 30 (5)
Optimizing Contractor Selection for Construction Packages in Capital Projects
In capital construction projects, the contractor selection process is executed in the front-end planning phase based on the experience and judgment of the project leadership team, whose members are considered experts. A comprehensive construction value packaging system (CVPS) is presented for assisting these project leaders in improving the efficacy, efficiency, and transparency of this process. The CVPS comprises an interface-oriented approach to work decomposition and a multicriteria approach to contractor selection. Selection is supported by software for iteratively computing and interactively visualizing the Pareto front of optimal work packages. The CVPS is validated by functional demonstration and a case study of a piping installation project in an energy infrastructure facility. The initial Pareto front had more than 100,000 optimal work packages representing different trade-offs between cost, time, contractor experience, maintenance rates, and financial stability. An iterative and interact...
- 1 September 2016
- Published by American Society of Civil Engineers (ASCE) in Journal of Computing in Civil Engineering
- Vol. 30 (5)
Optimizing Contractor Selection for Construction Packages in Capital Projects
In capital construction projects, the contractor selection process is executed in the front-end planning phase based on the experience and judgment of the project leadership team, whose members are considered experts. A comprehensive construction value packaging system (CVPS) is presented for assisting these project leaders in improving the efficacy, efficiency, and transparency of this process. The CVPS comprises an interface-oriented approach to work decomposition and a multicriteria approach to contractor selection. Selection is supported by software for iteratively computing and interactively visualizing the Pareto front of optimal work packages. The CVPS is validated by functional demonstration and a case study of a piping installation project in an energy infrastructure facility. The initial Pareto front had more than 100,000 optimal work packages representing different trade-offs between cost, time, contractor experience, maintenance rates, and financial stability. An iterative and interact...