Zeus: A system description of the two‐time winner of the collegiate SAE autodrive competition

Burnett, Keenan; Qian, Jingxing; Du, Xintong; Liu, Linqiao; Yoon, David J.; Shen, Tianchang; Sun, Susan; Samavi, Sepehr; Sorocky, Michael J.; Bianchi, Mollie; Zhang, Kaicheng; Arkhangorodsky, Arkady; Sykora, Quinlan; Lu, Shichen; Huang, Yizhou; Schoellig, Angela P.; Barfoot, Timothy D.

doi:10.1002/rob.21958

Cited by 20 publications

(18 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The red lines are the projected lane boundaries from the semantic map. Our approach is able to self-calibrate for the GPS-to-map offset and achieve alignment between the observed lane markings and the projected lane boundaries [4].…”

Section: Background and Motivationmentioning

confidence: 99%

“…As an illustrative example, aUToronto, the team that won the self-driving competition hosted by SAE International in 2019 [4], experienced an uncalibrated GPSto-map offset in the magnitude of a few metres, which was corrected manually just in time for the competition run, see Figure 1.1.…”

Section: Background and Motivationmentioning

confidence: 99%

“…However, as experienced by aUToronto, manual GPS calibration is often unreliable, and can lead to localization failures [4].…”

Section: Semantic Localizationmentioning

confidence: 99%

“…Table 5.2 summarizes the datasets with semantic maps and the components they lack for our experiments. Therefore, to demonstrate real-world feasibility, experiments were conducted using an internal dataset provided by aUToronto that was collected during the SAE AutoDrive competition at Mcity, where the incident of uncalibrated GPS occurred [4]. However, due to the lack of localization ground truth in the dataset for comparison, this will only serve as anecdotal results to verify the effectiveness of our approach.…”

Section: Mcity Experimentsmentioning

confidence: 99%

See 3 more Smart Citations

Self-Calibration of the Offset Between GPS and Semantic Map Frames for Robust Localization

Tseng

Schoellig

Barfoot

2021

2021 18th Conference on Robots and Vision (CRV)

Self Cite

View full text Add to dashboard Cite

In self-driving, GPS is generally considered to have insufficient positioning accuracy to stay in lane. Instead, many turn to LIDAR localization, but building and maintaining LIDAR maps can be costly. Another possibility is to use semantic cues such as lane markings and traffic lights to achieve localization, but these are usually not continuously visible. This can be remedied by combining semantic cues with GPS to fill in the gaps. However, due to biases accumulated between mapping and localization, the live GPS frame can be offset from the semantic map frame, requiring calibration.In this thesis, we propose a robust semantic localization algorithm that self-calibrates for the GPS-to-map offset by exploiting common semantic cues. We formulate the problem using a modified Iterated Extended Kalman Filter, which incorporates GPS and camera images for semantic cue detection. Experimental results show that the proposed algorithm achieves decimetre-level accuracy and is robust against sparse semantic features and frequent GPS dropouts.I also want to thank the Dean's Strategic Fund and NSERC for funding my project.Beyond that, I really appreciate the part-time opportunity offered by Tim, which helped address my financial concerns during my studies.Most importantly, I would not have made it this far without the unfailing support of my mother. Your words of encouragement and unconditional love is the reason I am able to overcome all the challenges and complete the degree.iiiA 6 × 6 adjoint of a transformation matrix (member of Ad(SE(3)))xii

show abstract

Section: Background and Motivationmentioning

confidence: 99%

Section: Background and Motivationmentioning

confidence: 99%

“…However, as experienced by aUToronto, manual GPS calibration is often unreliable, and can lead to localization failures [4].…”

Section: Semantic Localizationmentioning

confidence: 99%

Section: Mcity Experimentsmentioning

confidence: 99%

See 2 more Smart Citations

Self-Calibration of the Offset Between GPS and Semantic Map Frames for Robust Localization

Tseng

Schoellig

Barfoot

2021

2021 18th Conference on Robots and Vision (CRV)

Self Cite

View full text Add to dashboard Cite

show abstract

“…While the reliability of autonomous vehicles continues to improve, operating in rain and snow remains a challenge. Most autonomous driving systems rely primarily on a combination of cameras and lidar for perception, with radar sensors taking a back-seat role [13]. Cen and Newman [15] presented a successful application of a scanning radar sensor to largescale outdoor ego-motion estimation.…”

Section: Introductionmentioning

confidence: 99%

Radar Odometry Combining Probabilistic Estimation and Unsupervised Feature Learning

Burnett

Yoon

Schoellig

et al. 2021

Robotics: Science and Systems XVII

View full text Add to dashboard Cite

This paper presents a radar odometry method that combines probabilistic trajectory estimation and deep learned features without needing groundtruth pose information. The feature network is trained unsupervised, using only the on-board radar data. With its theoretical foundation based on a data likelihood objective, our method leverages a deep network for processing rich radar data, and a non-differentiable classic estimator for probabilistic inference. We provide extensive experimental results on both the publicly available Oxford Radar RobotCar Dataset and an additional 100 km of driving collected in an urban setting. Our sliding-window implementation of radar odometry outperforms most hand-crafted methods and approaches the current state of the art without requiring a groundtruth trajectory for training. We also demonstrate the effectiveness of radar odometry under adverse weather conditions. Code for this project can be found at: https://github.com/utiasASRL/hero radar odometry

show abstract

ForzaETH Race Stack—Scaled Autonomous Head‐to‐Head Racing on Fully Commercial Off‐the‐Shelf Hardware

Baumann,

Ghignone,

Kühne

et al. 2024

Journal of Field Robotics

View full text Add to dashboard Cite

Autonomous racing in robotics combines high‐speed dynamics with the necessity for reliability and real‐time decision‐making. While such racing pushes software and hardware to their limits, many existing full‐system solutions necessitate complex, custom hardware and software, and usually focus on Time‐TrIals rather than full unrestricted Head‐to‐head racing, due to financial and safety constraints. This limits their reproducibility, making advancements and replication feasible mostly for well‐resourced laboratories with comprehensive expertise in mechanical, electrical, and robotics fields. Researchers interested in the autonomy domain but with only partial experience in one of these fields, need to spend significant time with familiarization and integration. The ForzaETH Race Stack addresses this gap by providing an autonomous racing software platform designed for F1TENTH, a 1:10 scaled Head‐to‐Head autonomous racing competition, which simplifies replication by using commercial off‐the‐shelf hardware. This approach enhances the competitive aspect of autonomous racing and provides an accessible platform for research and development in the field. The ForzaETH Race Stack is designed with modularity and operational ease of use in mind, allowing customization and adaptability to various environmental conditions, such as track friction and layout, which is exemplified by the various modularly implemented state estimation and control systems. Capable of handling both Time‐Trials and Head‐to‐Head racing, the stack has demonstrated its effectiveness, robustness, and adaptability in the field by winning the official F1TENTH international competition multiple times. Furthermore, the stack demonstrated its reliability and performance at unprecedented scales, up to over on tracks up to 150 m in length.

show abstract

Zeus: A system description of the two‐time winner of the collegiate SAE autodrive competition

Cited by 20 publications

References 44 publications

Self-Calibration of the Offset Between GPS and Semantic Map Frames for Robust Localization

Self-Calibration of the Offset Between GPS and Semantic Map Frames for Robust Localization

Radar Odometry Combining Probabilistic Estimation and Unsupervised Feature Learning

ForzaETH Race Stack—Scaled Autonomous Head‐to‐Head Racing on Fully Commercial Off‐the‐Shelf Hardware

Contact Info

Product

Resources

About