Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
AI Powered Awareness for Increased Traffic Safety
Ansvarlig organisasjon
2022 (engelsk)Rapport (Annet vitenskapelig)
Abstract [sv]

The project: “AI Powered Awareness for Increased Traffic Safety”, has explored how multiple traffic- and road-related datasets can be analyzed using AI algorithms to identify roads with an elevated accident risk. Massive real-time data streams as well as large historic datasets have been evaluated by the project and then used to feed the AI algorithm. The ability to turn huge, but currently under-utilized, datasets into specific and quantified accident risk predictions may be used for proactive traffic management that ultimately could prevent accidents to occur. This would be a huge gain in the efforts to create a safer traffic system and the project has laid the foundation for this. Vehicle companies, map, and location data providers as well as traffic management authorities have been collaborating in the project to provide their knowledge, innovative concepts, and technology. Volvo Cars together with Zenseact created a data fusion platform that applies an AI algorithm for continuous accident risk evaluation on a road segment level. HERE made their “HERE Platform” available for the project, used both as a plentiful data source for Volvo Cars/Zenseact’s AI algorithm as well as a communication layer to forward predicted accident risks to a Central Traffic Control (CTC) cloud. The CTC cloud was provided by Carmenta using the Carmenta TrafficWatchTM platform with the purpose to collect and compile traffic situation data from various online sources and from a central level dispatch relevant traffic hazard warnings to connected organizations. The accident risks were transmitted from the CTC cloud to the Drive Sweden Innovation Cloud managed by Ericsson. The Innovation Cloud is a platform for business-to-business data and insights sharing, thus adding even more options for other organizations to share the accident risk predictions originating in the Volvo Cars/Zenseact data fusion platform. The Swedish Transport Administration and City of Gothenburg made huge contributions throughout the project, adding the traffic management authority perspective. At a workshop, traffic safety experts from Swedish Transport Administration, presented the latest findings and current guiding policies around traffic safety issues in Sweden and elaborated around Sweden’s “Vision Zero” initiative. Very important input for assessing how project findings and results can benefit a more pro-active traffic management. The project has successfully met its goals and not only implemented a first version of an AI-powered data fusion platform but also established and tested an end-to-end data flow mechanism to make the accident risk predictions immediately available to many connected organizations. All project members sees great potential in continuing the exploration of predictive awareness for increased traffic safety within their own organizations. Most likely the successful outcome of the “AI Powered Awareness for Increased Traffic Safety” project will lead to future follow-up Projects.

sted, utgiver, år, opplag, sider
AI Aware , 2022. , s. 16
Serie
Trafikverkets forskningsportföljer
Emneord [sv]
AI, Artificell intelligens, Människan i trafiken, Säkerhet och trygghet, Trafiksäkerhet, Vägar, Vägtrafik
HSV kategori
Forskningsprogram
FOI-portföljer, Strategiska initiativ
Identifikatorer
URN: urn:nbn:se:trafikverket:diva-17727Arkivnummer: TRV 2021/22663OAI: oai:DiVA.org:trafikverket-17727DiVA, id: diva2:1902213
Prosjekter
AI-förstärkt lägesbild för ökad trafiksäkerhet
Forskningsfinansiär
Swedish Transport Administration, TRV 2021/22663Tilgjengelig fra: 2024-10-01 Laget: 2024-10-01 Sist oppdatert: 2025-09-04bibliografisk kontrollert

Open Access i DiVA

AI Powered Awareness for Increased Traffic Safety(2061 kB)39 nedlastinger
Filinformasjon
Fil FULLTEXT01.pdfFilstørrelse 2061 kBChecksum SHA-512
ddf04187dfce057c9867acf881770560ee4e619da96b42e8a1f41ba272e92b46c26b11ea7ff1116963ffec78f64f513561428f967d0b318199cc7a754c1f201e
Type fulltextMimetype application/pdf

Søk utenfor DiVA

GoogleGoogle Scholar
Totalt: 39 nedlastinger
Antall nedlastinger er summen av alle nedlastinger av alle fulltekster. Det kan for eksempel være tidligere versjoner som er ikke lenger tilgjengelige

urn-nbn

Altmetric

urn-nbn
Totalt: 122 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
v. 2.47.0