2020
DOI: 10.1007/978-3-030-50086-3_1
|View full text |Cite
|
Sign up to set email alerts
|

Strategy Synthesis for Autonomous Driving in a Moving Block Railway System with Uppaal Stratego

Abstract: Moving block railway systems are the next generation signalling systems currently under development as part of the Shift2Rail European initiative, including autonomous driving technologies. In this paper, we model a suitable abstraction of a moving block signalling system with autonomous driving as a stochastic priced timed game. We then synthesise safe and optimal driving strategies for the model by applying advanced techniques that combine statistical model checking with reinforcement learning as provided by… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
11
0

Year Published

2020
2020
2024
2024

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 23 publications
(11 citation statements)
references
References 42 publications
(58 reference statements)
0
11
0
Order By: Relevance
“…This is currently investigated, with the application of stochastic model checking on a world model covering the operational design domain, as well as the trains and their ATP mechanisms discussed here. Moreover, the automated synthesis of safety supervisors from ATP-submodels of the world model will be explored with a novel methodological approach by Gleirscher et al [9], complementing existing results [1]. For calculating the probabilities of residual perceptor errors and for verifying stochastic independence between channels, we have developed a new method based on statistical tests, algorithms for the explanation of image classification results, and the construction of equivalence classes; the effectiveness of this method will be evaluated.…”
Section: Discussionmentioning
confidence: 99%
See 2 more Smart Citations
“…This is currently investigated, with the application of stochastic model checking on a world model covering the operational design domain, as well as the trains and their ATP mechanisms discussed here. Moreover, the automated synthesis of safety supervisors from ATP-submodels of the world model will be explored with a novel methodological approach by Gleirscher et al [9], complementing existing results [1]. For calculating the probabilities of residual perceptor errors and for verifying stochastic independence between channels, we have developed a new method based on statistical tests, algorithms for the explanation of image classification results, and the construction of equivalence classes; the effectiveness of this method will be evaluated.…”
Section: Discussionmentioning
confidence: 99%
“…Interventions are executed by the kernel through access to the train interface unit, for activating or releasing the service brakes or emergency brakes. The decisions about interventions are taken by the kernel based on the information provided by peripheral modules: (1) The odometry module and balise transmission module provide information for extracting trustworthy values for the actual train positions. As known from modern high-speed trains, additional positioning subsystems provide satellite positioning information in combination with radar sensor information to improve the precision and the reliability of the estimated train location.…”
Section: A Reference Architecture For Autonomous Train Controllersmentioning
confidence: 99%
See 1 more Smart Citation
“…Offline timed games [1] have been successfully applied to, e.g., railway, cruise, and traffic light control problems [12][17] [3] or power management [8]. These offline timed games assume complete knowledge of the behavior of a system and its environment, but few works exist that relax the requirement of complete information.…”
Section: Related Workmentioning
confidence: 99%
“…Medical needle steering describes the task of steering flexible and beveled needles in soft tissue towards a target e.g., for a biopsy or brachytherapy, while avoiding critical tissue layers and organs. The underlying problem is thus the safe navigation of a controllable entity through a (partially known and uncontrollable) environment in R 3 . The following requirements arise for a model intended to approach this underlying problem:…”
Section: Introductionmentioning
confidence: 99%