AI · SEMICONDUCTOR · FACT CHECK

SKD WIRE

EN WIRE · 2026-10-27 11:00

DGIST develops AI model that improves pedestrian walking-path prediction accuracy

True

DGIST develops AI model that improves pedestrian walking-path prediction accuracy
Image source: 조사 출처

The Daegu Gyeongbuk Institute of Science and Technology (DGIST) has developed an AI model that raises the accuracy of predicting pedestrian movement, according to a report carried by etnews and republished on news CDNs.

원문 주장 (KR)
DGIST, 보행 예측 정확도 높이는 AI 개발

Pedestrian prediction for safer autonomous systems

The research targets pedestrian trajectory prediction, a core capability for autonomous vehicles, service robots and other AI systems that must anticipate where people will move next. The DGIST-developed AI improves the accuracy of such predictions, per the report.

The images accompanying the coverage were published alongside the article, including materials hosted on the etnews image CDN and aitimes.kr.

Details of the method remain limited in the published coverage

The available coverage confirms the outcome — a DGIST AI development that boosts pedestrian prediction accuracy — but does not provide further technical specifics, benchmark figures or funding details. Readers seeking the full methodology would need to consult the institute's original announcement or the underlying publication.

The claim as circulated — that DGIST has developed an AI that raises pedestrian prediction accuracy — is judged True.

Verdict: True

Sources — primary documents (9)
  1. https://www.dgist.ac.kr/bbs/BBSMSTR_000000000080/view.do?nttId=B000000112405Tb6sU3
  2. https://pubmed.ncbi.nlm.nih.gov/42838425/
  3. https://www.aitimes.kr/news/articleView.html?idxno=42439
  4. https://www.etnews.com/20261027000020
  5. https://www.sciencedirect.com/science/article/pii/S1053811926005951
  6. https://www.dgist.ac.kr/cmm/fms/FileDown.do?atchFileId=FILE_000000171226Fn1&fileSn=0/1/2
  7. https://img.etnews.com/news/article/2026/10/27/news-p.v1.20261027.6773653b65dd4fecbf04dbf741bce0bd_P1.jpg
  8. https://img.etnews.com/news/article/2026/10/27/news-p.v1.20261027.474fca9e319541ecbf2e7cfb8cf42001_P1.png
  9. https://cdn.aitimes.kr/news/photo/202610/42439_63794_4845.png

KR: /news/20261027-af6a1c · 판정: 사실