A Perceptually-Validated Metric for Crowd Trajectory Quality Evaluation

Simulating crowds requires controlling a very large number of trajectories and is usually performed using crowd motion algorithms for which appropriate parameter values need to be found. The study of the relation between parametric values for simulation techniques and the quality of the resulting trajectories has been studied either through perceptual experiments or by comparison with real crowd trajectories. In this paper, we integrate both strategies. A quality metric, QF, is proposed to abstract from reference data while capturing the most salient features that affect the perception of trajectory realism. QF weights and combines cost functions that are based on several individual, local and global properties of trajectories. These trajectory features are selected from the literature and from interviews with experts. To validate the capacity of QF to capture perceived trajectory quality, we conduct an online experiment that demonstrates the high agreement between the automatic quality score and non-expert users. To further demonstrate the usefulness of QF, we use it in a data-free parameter tuning application able to tune any parametric microscopic crowd simulation model that outputs independent trajectories for characters. The learnt parameters for the tuned crowd motion model maintain the influence of the reference data which was used to weight the terms of QF.

Small Flying Object Classifications Based on Trajectories and Support Vector Machines

This paper presents an alternative approach to identify and classify the group of small flying objects especially drones from others, notably birds and kites (inclusive of kiteflying), in near field, by examining the pattern of their flight paths and trajectories. The trajectories of the drones and other flying objects were extracted from multiple clips of videos including various natural and synthetic database. Four trajectories characteristics are observed and extracted from the object’s flight paths, i.e., heading or turning angle, curvature, pace velocity, and pace acceleration. Subsequently, principal component analyses were applied to reduce the number of these trajectory features from 4 to 2 parameters. Multi-class classification by support vector machine (SVM) with non-linear transformation kernel was used. Multiple classification models were developed by several algorithms with various transformation kernels. The hyperparameters were optimized using Bayesian optimization. The performances of the different models are compared through the prediction accuracy of the test data.

A non-intrusive load identification algorithm based on deep learning and a compound feature

Aiming at the limitations of using a single feature for load identification, a non-intrusive load identification algorithm based on deep learning and compound features is proposed. The pixelated V-I trajectory characteristics and current harmonic characteristics are extracted by analyzing the load data under high-frequency sampling. Using the feature extraction capabilities of neural networks, the combination of pixelated V-I trajectory features and current harmonic features is realized. Finally, the composite feature is used as the new load feature to train the neural network for non-invasive load identification. The experimental results show that the two-layer neural network constructed by the algorithm can take advantage of the complementarity between the two features, thereby improving the load identification ability.

An Air Traffic Controller Action Extraction-Prediction Model Using Machine Learning Approach

In air traffic control, the airspace is divided into several smaller sectors for better management of air traffic and air traffic controller workload. Such sectors are usually managed by a team of two air traffic controllers: planning controller (D-side) and executive controller (R-side). D-side controller is responsible for processing flight-plan information to plan and organize the flow of traffic entering the sector. R-side controller deals with ensuring safety of flights in their sector. A better understanding and predictability of D-side controller actions, for a given traffic scenario, may help in automating some of its tasks and hence reduce workload. In this paper, we propose a learning model to predict D-side controller actions. The learning problem is modeled as a supervised learning problem, where the target variables are D-side controller actions and the explanatory variables are the aircraft 4D trajectory features. The model is trained on six months of ADS-B data over an en-route sector, and its generalization performance was assessed, using crossvalidation, on the same sector. Results indicate that the model for vertical maneuver actions provides highest prediction accuracy (99%). Besides, the model for speed change and course change action provides predictability accuracy of 80% and 87%, respectively. The model to predict the set of all the actions (altitude, speed, and course change) for each flight achieves an accuracy of 70% implying for 70% of flights; D-side controller’s action can be predicted from trajectory information at sector entry position. In terms of operational validation, the proposed approach is envisioned as ATCO assisting tool, not an autonomous tool. Thus, there is always ATCO discretion element, and as more ATCO actions are collected, the models can be further trained for better accuracy. For future work, we will consider expanding the feature set by including parameters such as weather and wind. Moreover, human in the loop simulation will be performed to measure the effectiveness of the proposed approach.

Predictive Maps in Rats and Humans for Spatial Navigation

Much of our understanding of navigation has come from the study of rats, humans and simulated artificial agents. To date little attempt has been made to integrate these approaches into a common framework to understand mechanisms that may be shared across mammals and the extent to which different instantiations of agents best capture mammalian navigation behaviour. Here, we report a comparison of rats, humans and reinforcement learning (RL) agents in a novel open-field navigation task (‘Tartarus Maze’) requiring dynamic adaptation (shortcuts and detours) to changing obstructions in the path to the goal. We find humans and rats are remarkably similar in patterns of choice in the task. The patterns in their choices, dwell maps and changes over time reveal that both species show the greatest similarity to RL agents utilising a predictive map: the successor representation. Humans also display trajectory features similar to a model-based RL agent. Our findings have implications for models seeking to explain mammalian navigation in dynamic environments and highlight the utility of modelling the behaviour of different species in the same frame-work in comparison to RL agents to uncover the potential mechanisms used for behaviour.

Filipino and Indonesian Migrant Domestic Workers in Hong Kong: Their Life Courses in Migration

This article presents a portrayal of Filipino and Indonesian female domestic workers’ life courses in migration, using the life history calendar data from the 2017 survey of migrant domestic workers in Hong Kong. Applying sequence analysis, we first analyzed migration trajectory features such as individual migration trajectories, duration spent in each state, and longitudinal diversity of state distributions. We found that Indonesian domestic workers, compared with their Filipino counterparts, are more diverse in their migration histories, indicating involvements in serial migration. We also conducted a cluster analysis of the domestic workers’ migratory trajectories. The analysis yielded three meaningful clusters/types of migrant workers—those moved late in life, those who participated in serial migration, and those migrated directly from their home country to Hong Kong. Finally, we investigated the effect of a complex migration history on job satisfaction and the characteristics of membership in the three ideal-typical migration types among the domestic workers older than 39 years.