MuRAL is a dataset of activities of daily living in a multi-resident setting, annotated with fine-grained natural language descriptions, resident identities, and high-level activity labels. This dataset was collected in the Domus intelligent apartment platform, in 2025.
This dataset was collected to support future research on multi-resident activity recognition, in particular LLM-powered approaches that can leverage common-knowledge context reasoning to infer activities, using natural language descriptions of the environment and sensors installed in it.
This dataset is openly available, simply contact me by email to ask for access to the dataset. (a download link will be available soon)
Any use of this dataset in publications must cite the following publication:
Xi Chen, Julien Cumin, Fano Ramparany, Dominique Vaufreydaz. MuRAL: a multi-resident ambient sensor dataset annotated with natural language for activities of daily living. arXiv preprint 2025.
Orange4Home is a dataset of routines of daily living captured in Amiqual4Home’s smart home environment, from January 30th, 2017 to February 24th, 2017.
This dataset is the result of a joint work between Orange and Inria. The experiment was conducted by Julien Cumin, Grégoire Lefebvre, Fano Ramparany and James L. Crowley, with technical and organisational help from Nicolas Bonnefond and Stan Borkowski
More information about the dataset is available on Amiqual4Home’s website.
This dataset is openly available and can be downloaded here
Any use of this dataset in publications must cite the following publication:
Julien Cumin, Grégoire Lefebvre, Fano Ramparany, James L. Crowley. A dataset of routine daily activities in an instrumented home. In 11th international conference on ubiquitous computing and ambient intelligence (UCAmI), 2017.