A Recommender System for Increasing Energy Efficiency of Solar-Powered Smart Homes

Photovoltaic installations can be environmentally beneficial to a greater or lesser extent, depending on the conditions. If the energy produced is not used, it is redirected to the grid, otherwise a battery with a high ecological footprint is needed to store it. To alleviate this problem, an innovative recommender system is proposed for residents of smart homes equipped with battery-free solar panels to optimise the energy produced. Using artificial intelligence, the system is designed to predict the energy produced and consumed for the day ahead using three data sources: sensor logs from the home automation solution, data collected by the solar inverter, and weather data. Based on these predictions, recommendations are then generated and ranked by relevance. Data collected over 76 days were used to train two variants of the system, considering or without considering energy consumption. Recommendations selected by the system over 14 days were randomly picked to be evaluated for relevance, ranking, and diversity by 11 people. The results show that it is difficult to predict residents' consumption based solely on sensor logs. On average, respondents reported that 74% of the recommendations were relevant, while the values contained in them (i.e., accuracy of times of day and kW energy) were accurate in 66% (variant 1) and 77% of cases (variant 2). Also, the ranking of the recommendations was considered logical in 91% and 88% of cases. Overall, residents of such solar-powered smart homes might be willing to use such a system to optimise the energy produced. However, further research should be conducted to improve the accuracy of the values contained in the recommendations.

A dataset on the physiological state and behavior of drivers in conditionally automated driving

This dataset contains data of 346 drivers collected during six experiments conducted in a fixed-base driving simulator. Five studies simulated conditionally automated driving (L3-SAE), and the other one simulated manual driving (L0-SAE). The dataset includes physiological data (electrocardiogram (ECG), electrodermal activity (EDA), and respiration (RESP)), driving and behavioral data (reaction time, steering wheel angle, …), performance data of non-driving-related tasks, and questionnaire responses. Among them, measures from standardized questionnaires were collected, either to control the experimental manipulation of the driver's state, or to measure constructs related to human factors and driving safety (drowsiness, mental workload, affective state, situation awareness, situational trust, user experience). In the provided dataset, some raw data have been processed, notably physiological data from which physiological indicators (or features) have been calculated. The latter can be used as input for machine learning models to predict various states (sleep deprivation, high mental workload, ...) that may be critical for driver safety. Subjective self-reported measures can also be used as ground truth to apply regression techniques. Besides that, statistical analyses can be performed using the dataset, in particular to analyze the situational awareness or the takeover quality of drivers, in different states and different driving scenarios. Overall, this dataset contributes to better understanding and consideration of the driver's state and behavior in conditionally automated driving. In addition, this dataset stimulates and inspires research in the fields of physiological/affective computing and human factors in transportation, and allows companies from the automotive industry to better design adapted human-vehicle interfaces for safe use of automated vehicles on the roads.

Carrying a passenger and relaxation before driving: Classification of young drivers' physiological activation

Drivers are often held responsible for road crashes. Previous research has shown that stressors such as carrying passengers in the vehicle can be a source of accidents for young drivers. To mitigate this problem, this study investigated whether the presence of a passenger behind the wheel can be predicted using machine learning, based on physiological signals. It also addresses the question whether relaxation before driving can positively influence the driver's state and help controlling the potential negative consequences of stressors. Sixty young participants completed a 10‐min driving simulator session, either alone or with a passenger. Before their driving session, participants spent 10 min relaxing or listening to an audiobook. Physiological signals were recorded throughout the experiment. Results show that drivers experience a higher increase in skin conductance when driving with a passenger, which can be predicted with 90%‐accuracy by a k‐nearest neighbors classifier. This might be a possible explanation for increased risk taking in this age group. Besides, the practice of relaxation can be predicted with 80% accuracy using a neural network. According to the statistical analysis, the potential beneficial effect of relaxation did not carry out on the driver's physiological state while driving, although machine learning techniques revealed that participants who exercised relaxation before driving could be recognized with 70% accuracy. Analysis of physiological characteristics after classification revealed several relevant physiological indicators associated with the presence of a passenger and relaxation.

Non-Intrusive Contact Respiratory Sensor for Vehicles

In this work, we propose a low-cost solution capable of collecting the driver's respiratory signal in a robust and non-intrusive way by contact with the chest and abdomen. It consists of a microcontroller and two piezoelectric sensors with their respective 3D printed plastic housings attached to the seat belt. An iterative process was conducted to find the optimal shape of the sensor housing. The location of the sensors can be easily adapted by sliding them along the seat belt. A few participants took part in three test sessions in a driving simulator. They had to perform various activities: resting, deep breathing, manual driving, and a non-driving-related task during automated driving. The subjects' breathing rates were calculated from raw data collected with a reference chest belt, each sensor alone, and the fusion of the two. Results indicate that respiratory rate could be assessed from a single sensor located on the chest with an average absolute error of 0.92 min^-1 across all periods, dropping to 0.13 min^-1 during deep breathing. Sensor fusion did not improve system performance. A 4-pole filter with a cutoff frequency of 1 Hz emerged as the best option to minimize the error during the different periods. The results suggest that such a system could be used to assess the driver's breathing rate while performing various activities in a vehicle.

Relevant Physiological Indicators for Assessing Workload in Conditionally Automated Driving, Through Three-Class Classification and Regression

In future conditionally automated driving, drivers may be asked to take over control of the car while it is driving autonomously. Performing a non-driving-related task could degrade their takeover performance, which could be detected by continuous assessment of drivers' mental load. In this regard, three physiological signals from 80 subjects were collected during 1 h of conditionally automated driving in a simulator. Participants were asked to perform a non-driving cognitive task (N-back) for 90 s, 15 times during driving. The modality and difficulty of the task were experimentally manipulated. The experiment yielded a dataset of drivers' physiological indicators during the task sequences, which was used to predict drivers' workload. This was done by classifying task difficulty (three classes) and regressing participants' reported level of subjective workload after each task (on a 0–20 scale). Classification of task modality was also studied. For each task, the effect of sensor fusion and task performance were studied. The implemented pipeline consisted of a repeated cross validation approach with grid search applied to three machine learning algorithms. The results showed that three different levels of mental load could be classified with a f1-score of 0.713 using the skin conductance and respiration signals as inputs of a random forest classifier. The best regression model predicted the subjective level of workload with a mean absolute error of 3.195 using the three signals. The accuracy of the model increased with participants' task performance. However, classification of task modality (visual or auditory) was not successful. Some physiological indicators such as estimates of respiratory sinus arrhythmia, respiratory amplitude, and temporal indices of heart rate variability were found to be relevant measures of mental workload. Their use should be preferred for ongoing assessment of driver workload in automated driving.

The interplay between task difficulty and microsaccade rate: Evidence for the critical role of visual load

In previous research, microsaccades have been suggested as psychophysiological indicators of task load. So far, it is still under debate how different types of task demands are influencing microsaccade rate. This piece of research examines the relation between visual load, mental load and microsaccade rate. Fourteen participants carried out a continuous performance task (n-back), in which visual (letters vs. abstract figures) and mental task load (1-back to 4-back) were manipulated as within-subjects variables. Eye tracking data, performance data as well as subjective workload were recorded. Data analysis revealed an increased level of microsaccade rate for stimuli of high visual demand (i.e. abstract figures), while mental demand (n-back-level) did not modulate microsaccade rate. In conclusion, the present results suggest that microsaccade rate reflects visual load of a task rather than its mental load.

Classification of Drivers' Workload Using Physiological Signals in Conditional Automation

The use of automation in cars is increasing. In future vehicles, drivers will no longer be in charge of the main driving task and may be allowed to perform a secondary task. However, they might be requested to regain control of the car if a hazardous situation occurs (i.e., conditionally automated driving). Performing a secondary task might increase drivers' mental workload and consequently decrease the takeover performance if the workload level exceeds a certain threshold. Knowledge about the driver's mental state might hence be useful for increasing safety in conditionally automated vehicles. Measuring drivers' workload continuously is essential to support the driver and hence limit the number of accidents in takeover situations. This goal can be achieved using machine learning techniques to evaluate and classify the drivers' workload in real-time. To evaluate the usefulness of physiological data as an indicator for workload in conditionally automated driving, three physiological signals from 90 subjects were collected during 25 min of automated driving in a fixed-base simulator. Half of the participants performed a verbal cognitive task to induce mental workload while the other half only had to monitor the environment of the car. Three classifiers, sensor fusion and levels of data segmentation were compared. Results show that the best model was able to successfully classify the condition of the driver with an accuracy of 95%. In some cases, the model benefited from sensors' fusion. Increasing the segmentation level (e.g., size of the time window to compute physiological indicators) increased the performance of the model for windows smaller than 4 min, but decreased for windows larger than 4 min. In conclusion, the study showed that a high level of drivers' mental workload can be accurately detected while driving in conditional automation based on 4-min recordings of respiration and skin conductance.