Ensemble Bayesian Decision Making with Redundant Deep Perceptual Control Policies

By Keuntaek Lee, Ziyi Wang, Bogdan I. Vlahov, Harleen K. Brar, Evangelos A. Theodorou. 2018

32 citations recorded by the source corpus.

This work presents a novel ensemble of Bayesian Neural Networks for control of safety-critical systems. Decision making for safety-critical systems is challenging due to performance requirements with significant consequences in the event of failure. In practice, failure of such systems can be avoided by introducing redundancies of control. Neural Networks are generally not used for safety-critical systems as they can behave in unexpected ways in response to novel inputs. In addition, there may not be any indication as to when they will fail. BNNs have been recognized for their ability to produce not only viable outputs but also provide a measure of uncertainty in these outputs. This work combines the knowledge of prediction uncertainty obtained from BNNs and ensemble control for a redundant control methodology. Our technique is applied to an agile autonomous driving task. Multiple BNNs are trained to control a vehicle in an end-to-end fashion on different sensor inputs provided by the system. We show that an individual network is successful in maneuvering around the track but crashes in the presence of unforeseen input noise. Our proposed ensemble of BNNs shows successful task performance even in the event of multiple sensor failures.

Subfields: Artificial Intelligence, Automotive Engineering, Reinforcement Learning, Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Davide Scaramuzza
2. Antonio Loquercio
3. Mattia Segu
4. Alexander Liniger
5. Maximilian Pichler
6. Florian Hartig

In Automotive Engineering

1. Davide Scaramuzza
2. Antonio Loquercio
3. Mattia Segu
4. Rahul Mangharam
5. Alexander Liniger

In Reinforcement Learning

1. Mattia Segu
2. Antonio Loquercio
3. Davide Scaramuzza
4. Alexander Liniger

In Probabilistic & Bayesian Methods

1. Mattia Segu
2. Antonio Loquercio
3. Davide Scaramuzza
4. Florian Hartig
5. Maximilian Pichler

Citations in the corpus

• A General Framework for Uncertainty Estimation in Deep Learning (2019, background): “[2] propose to exploit an ensemble of redundant neural networks trained on different sensors modalities, trusting at test time the prediction with lowest uncertainty.”
• Autonomous Vehicles on the Edge: A Survey on Autonomous Vehicle Racing (2022, background): “[115] 2021 Reinforcement Learning Vision based Planning Bayesian Deciscion Making Yes AutoRally”