<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>6539</id>
    <completedYear/>
    <publishedYear>2020</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>91</pageFirst>
    <pageLast>95</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <articleNumber/>
    <type>conferenceobject</type>
    <publisherName>RWTH Aachen</publisherName>
    <publisherPlace>Aachen</publisherPlace>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>2025-12-16</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">PURSS: Towards Perceptual Uncertainty Aware Responsibility Sensitive Safety&#13;
with ML</title>
    <abstract language="eng">Automated driving is an ML-intensive problem and its safety depends on the integrity of perception as well as planning and control. Responsibility Sensitive Safety (RSS) is a recent approach to promote safe planning and control that relies on perfect perception; however, perceptual uncertainty is always present, and this causes the possibility of misperceptions that can lead an autonomous vehicle to allow unsafe actions. In this position paper, we sketch a novel proposal for a formal model of perception coupled with RSS to help mitigate the impact of misperception by using information about perceptual uncertainty. The approach expresses uncertainty as imprecise perceptions that are consumed by RSS and cause it to limit actions to those that support safe behaviour given the perceptual uncertainty. We illustrate our approach using examples and discuss its implications and limitations.</abstract>
    <parentTitle language="deu">Proceedings of the Workshop on Artificial Intelligence Safety (SafeAI 2020), co-located with 34th AAAI Conference on Artificial Intelligence (AAAI 2020)</parentTitle>
    <identifier type="url">https://ceur-ws.org/Vol-2560/</identifier>
    <identifier type="urn">urn:nbn:de:bvb:573-65390</identifier>
    <enrichment key="THI_openaccess">ja</enrichment>
    <enrichment key="THI_review">peer-review</enrichment>
    <enrichment key="THI_conferenceName">AAAI Workshop on Artificial Intelligence Safety (SafeAI 2020), New York (USA), 07.02.2020</enrichment>
    <enrichment key="opus.doi.autoCreate">false</enrichment>
    <enrichment key="opus.urn.autoCreate">true</enrichment>
    <licence>Creative Commons BY 4.0</licence>
    <author>
      <first_name>Rick</first_name>
      <last_name>Salay</last_name>
    </author>
    <editor>
      <first_name>Huáscar</first_name>
      <last_name>Espinoza</last_name>
    </editor>
    <author>
      <first_name>Krzysztof</first_name>
      <last_name>Czarnecki</last_name>
    </author>
    <editor>
      <first_name>José</first_name>
      <last_name>Hernández-Orallo</last_name>
    </editor>
    <author>
      <first_name>Ignacio</first_name>
      <last_name>Alvarez</last_name>
    </author>
    <editor>
      <first_name>Xin Cynthia</first_name>
      <last_name>Chen</last_name>
    </editor>
    <author>
      <first_name>Maria</first_name>
      <last_name>Elli</last_name>
    </author>
    <editor>
      <first_name>Seán S.</first_name>
      <last_name>ÓhÉigeartaigh</last_name>
    </editor>
    <author>
      <first_name>Sean</first_name>
      <last_name>Sedwards</last_name>
    </author>
    <editor>
      <first_name>Xiaowei</first_name>
      <last_name>Huang</last_name>
    </editor>
    <author>
      <first_name>Jack</first_name>
      <last_name>Weast</last_name>
    </author>
    <editor>
      <first_name>Mauricio</first_name>
      <last_name>Castillo-Effen</last_name>
    </editor>
    <editor>
      <first_name>Richard</first_name>
      <last_name>Mallah</last_name>
    </editor>
    <editor>
      <first_name>John</first_name>
      <last_name>McDermid</last_name>
    </editor>
    <collection role="open_access" number="">open_access</collection>
    <collection role="persons" number="62036">Alvarez, Ignacio</collection>
    <thesisPublisher>Technische Hochschule Ingolstadt</thesisPublisher>
    <file>https://opus4.kobv.de/opus4-haw/files/6539/paper34.pdf</file>
  </doc>
</export-example>
