Predicting and Preventing Unsafe Situations for Service Robots

Summary:

One crucial concern is execution safety in domestic environments where robots work with humans and everyday objects. Unsafe situations can arise from the robot’s own actions, the actions of a human, or environmental factors. These situations should be avoided as much as possible. This project aims to design and develop an effective system for service robots to predict and prevent unsafe states.

This research is funded by a grant from the Scientific and Technological Research Council of Turkey (TUBITAK), Grant No. 119E-436.

Project Team:

PI: Dr. Sanem Sariel

Researchers: Dr. Eren Aksoy, Dr. Ayşe Tosun, A. Cihan Ak, Arda İnceoğlu, Gamze Akyol, Sadık Ugursoy, Ali Tolga Dincer

Dataset

Download Data (~9.5GB compressed, ~20GB decompressed)

Download Annotations

Sample executions from the dataset:

PlacePourPut-InPut-OnPush
Success
Failure

Publications

FINO-Net: A Deep Multimodal Sensor Fusion Framework for Manipulation Failure Detection
Arda Inceoglu, Eren Erdal Aksoy, Abdullah Cihan Ak, Sanem Sariel
[PDF] [Code and Data]