Modeling Human Activities Using Behaviour Trees in Smart Homes
Bibliographic record
Abstract
With the aging population, researchers around the world are investigating technological solutions to help seniors stay at home as long as possible. One of them is the concept of smart home, which is an intelligent house equipped with sensors and actuators. Aging people often suffers from physical and cognitive impairments, which limit their abilities to perform their Activities of Daily Living (ADL). Therefore, the smart home needs to be able to assist its resident in carrying out their ADL, when it is required. Recognising the ongoing ADL constitutes then a key challenge of the assistive services. Being able to simulate users' behaviour is also an important issue, as well as being able to find an assistive step-by-step solution when something goes wrong. However, all theses challenges need to rely on a knowledge base of activities' models. In the past, many researchers tried to make use of some logical encoding of the activities by exploiting, for instance, first order logic. These approaches work fine for the inferential process but they are very rigid, complex and time consuming. More recently, scientists in the field tried to represent the activities using stochastic models, such as Bayesian Networks or Markov Model. These probabilistic methods do not represent activities very naturally and are very static state-transition models. In this paper, we propose the use of Behaviour Trees (BT) as a means to represent the user's ADL in a smart home. BTs are mainly used in the video game industry as a powerful tool to model the behaviour of non-player characters. BTs allow the modelling of activities with a flexible, well-defined approach. We will present a first exploitation of the behaviour trees in a smart home simulator.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".