Handover Failure Detection dataset

This dataset contains multimodal data recorded on robots performing robot-to-human and human-to-robot handovers. The intended application of the dataset is to develop and benchmark methods for failure detection during the handovers. Thus the trials in the dataset contain both successful and failed handover actions. For a more detailed description of the dataset, please see the included Datasheet.

A description of the files:

• training_set.tar.gz, val_data.tar.gz, test_set.tar.gz: contains the video and other multimodal data from 589 trials split into training, validation and test sets
• sample_training_set.tar.gz: contains two sample trials from the training set included if you want to just download a sample of the dataset
• training_labels.tar.gz, val_labels.tar.gz, test_labels.tar.gz: contains the labels of the outcomes for all the trials
• wrench_stats.json: mean and standard deviation of the force-torque data from the training set
• class_names.json: names of the outcomes, human actions and robot actions as dictionaries
• datasheet.pdf: datasheet describing the motivation, contents, collection process, etc. of the dataset