Ensemble Bayesian Decision Making with Redundant Deep Perceptual Control Policies

Lee, Keuntaek; Wang, Ziyi; Vlahov, Bogdan; Brar, Harleen K.; Theodorou, Evangelos A.

doi:10.1109/icmla.2019.00145

Cited by 18 publications

(11 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The racing task provides a clear objective function (fastest lap time) for the algorithm training and the race track provides with its clear driveable area and one class of objects a perfect proving ground. Researchers in this field displayed partial end-toend approaches (Lee et al, 2019;Weiss & Behl, 2020) that combine DNNs with MPC methods to create and follow dynamic trajectories. In addition, by using algorithms from the field of RL (e.g., Soft-Actor-Critic and Q-Learning), researchers were able to demonstrate how to train an agent to drive fast (de Bruin et al, 2018;Jaritz et al, 2018), how to train an agent to overtake other agents on the race track (Song et al, 2021) and how to bridge the sim-to-real gap with model-based RL approaches (Brunnbauer et al, 2021).…”

Section: Softwarementioning

confidence: 99%

TUM autonomous motorsport: An autonomous racing software for the Indy Autonomous Challenge

Betz

Fent

et al. 2023

Journal of Field Robotics

View full text Add to dashboard Cite

For decades, motorsport has been an incubator for innovations in the automotive sector and brought forth systems, like, disk brakes or rearview mirrors. Autonomous racing series such as Roborace, F1Tenth, or the Indy Autonomous Challenge (IAC) are envisioned as playing a similar role within the autonomous vehicle sector, serving as a proving ground for new technology at the limits of the autonomous systems capabilities. This paper outlines the software stack and approach of the TUM Autonomous Motorsport team for their participation in the IAC, which holds two competitions: A single‐vehicle competition on the Indianapolis Motor Speedway and a passing competition at the Las Vegas Motor Speedway. Nine university teams used an identical vehicle platform: A modified Indy Lights chassis equipped with sensors, a computing platform, and actuators. All the teams developed different algorithms for object detection, localization, planning, prediction, and control of the race cars. The team from Technical University of Munich (TUM) placed first in Indianapolis and secured second place in Las Vegas. During the final of the passing competition, the TUM team reached speeds and accelerations close to the limit of the vehicle, peaking at around 2700.25em km h − 1 $270\,\text{km\hspace{0.05em}h}{}^{-1}$ and 280.25em m s − 2 $28\,ms{}^{-2}$. This paper will present details of the vehicle hardware platform, the developed algorithms, and the workflow to test and enhance the software applied during the 2‐year project. We derive deep insights into the autonomous vehicle's behavior at high speed and high acceleration by providing a detailed competition analysis. On the basis of this, we deduce a list of lessons learned and provide insights on promising areas of future work based on the real‐world evaluation of the displayed concepts.

show abstract

Section: Softwarementioning

confidence: 99%

TUM autonomous motorsport: An autonomous racing software for the Indy Autonomous Challenge

Betz

Fent

et al. 2023

Journal of Field Robotics

View full text Add to dashboard Cite

show abstract

“…Monte-Carlo sampling methods involve sampling the parameters from a distribution and are generally obtained using an ensemble of neural network predictions. The prediction ensemble could either be generated by differently trained networks [29], or by using dropout at test-time [28].…”

Section: Epistemic Uncertainty In Graph Neural Networkmentioning

confidence: 99%

A General Framework for quantifying Aleatoric and Epistemic uncertainty in Graph Neural Networks

Munikoti¹,

Agarwal²,

Das³

et al. 2022

Preprint

View full text Add to dashboard Cite

Graph Neural Networks (GNN) provide a powerful framework that elegantly integrates Graph theory with Machine learning for modeling and analysis of networked data. We consider the problem of quantifying the uncertainty in predictions of GNN stemming from modeling errors and measurement uncertainty. We consider aleatoric uncertainty in the form of probabilistic links and noise in feature vector of nodes, while epistemic uncertainty is incorporated via a probability distribution over the model parameters. We propose a unified approach to treat both sources of uncertainty in a Bayesian framework, where Assumed Density Filtering is used to quantify aleatoric uncertainty and Monte Carlo dropout captures uncertainty in model parameters. Finally, the two sources of uncertainty are aggregated to estimate the total uncertainty in predictions of a GNN. Results in the real-world datasets demonstrate that the Bayesian model performs at par with a frequentist model and provides additional information about predictions uncertainty that are sensitive to uncertainties in the data and model.

show abstract

“…However, for most neural-network architectures, the true posterior distribution over parameters is intractable and must be approximated. The most common approximation scheme seen in the literature is to use dropout at test time also called MCDropout [117,196], which was successfully applied to automonous driving [130], robot control [213], and health pronostics [170]. Given a specific input x, the variance of the model's predictions from different forward passes with random shutdown of neural activations is used as a measure of the epistemic uncertainty at x.…”

Section: Epistemic Uncertaintymentioning

confidence: 99%

How to Certify Machine Learning Based Safety-critical Systems? A Systematic Literature Review

Tambon,

Laberge,

et al. 2021

Preprint

View full text Add to dashboard Cite

Context: Machine Learning (ML) has been at the heart of many innovations over the past years. However, including it in so-called "safety-critical" systems such as automotive or aeronautic has proven to be very challenging, since the shift in paradigm that ML brings completely changes traditional certification approaches.Objective: This paper aims to elucidate challenges related to the certification of ML-based safety-critical systems, as well as the solutions that are proposed in the literature to tackle them, answering the question "How to Certify Machine Learning Based Safety-critical Systems?".Method: We conduct a Systematic Literature Review (SLR) of research papers published between 2015 to 2020, covering topics related to the certification of ML systems. In total, we identified 217 papers covering topics considered to be the main pillars of ML certification: Robustness, Uncertainty, Explainability, Verification, Safe Reinforcement Learning, and Direct Certification. We analyzed the main trends and problems of each sub-field and provided summaries of the papers extracted.Results: The SLR results highlighted the enthusiasm of the community for this subject, as well as the lack of diversity in term of datasets and type of ML models. It also emphasized the need to further develop connections between academia and industries to deepen the domain study. Finally, it also illustrated This work is supported by the DEEL project CRDPJ 537462-18 funded by the National Science and Engineering Research Council of Canada (NSERC) and the Consortium for Research and Innovation in Aerospace in Québec (CRIAQ), together with its industrial partners Thales Canada inc, Bell Textron Canada Limited, CAE inc and Bombardier inc.

show abstract

Ensemble Bayesian Decision Making with Redundant Deep Perceptual Control Policies

Cited by 18 publications

References 12 publications

TUM autonomous motorsport: An autonomous racing software for the Indy Autonomous Challenge

TUM autonomous motorsport: An autonomous racing software for the Indy Autonomous Challenge

A General Framework for quantifying Aleatoric and Epistemic uncertainty in Graph Neural Networks

How to Certify Machine Learning Based Safety-critical Systems? A Systematic Literature Review

Contact Info

Product

Resources

About