Deep learning for sensor-based activity recognition: A survey

With a widespread of various sensors embedded in mobile devices, the analysis of human daily activities becomes more common and straightforward. This task now arises in a range of applications such as healthcare monitoring, fitness tracking or user-adaptive systems, where a general model capable of instantaneous activity recognition of an arbitrary user is needed. In this paper, we present a user-independent deep learning-based approach for online human activity classification. We propose using Convolutional Neural Networks for local feature extraction together with simple statistical features that preserve information about the global form of time series. Furthermore, we investigate the impact of time series length on the recognition accuracy and limit it up to 1 s that makes possible continuous real-time activity classification. The accuracy of the proposed approach is evaluated on two commonly used WISDM and UCI datasets that contain labeled accelerometer data from 36 and 30 users respectively, and in cross-dataset experiment. The results show that the proposed model demonstrates state-of-the-art performance while requiring low computational cost and no manual feature engineering.

Real-time human activity recognition from accelerometer data using Convolutional Neural Networks

Sensor-based and vision-based human activity recognition: A comprehensive survey

Data fusion and multiple classifier systems for human activity detection and health monitoring: Review and open research directions

Human activity recognition (HAR) is currently recognized as a key element of a more general framework designed to perform continuous monitoring of human behaviors in the area of ambient assisted living (AAL), well-being management, medical diagnosis, elderly care, rehabilitation, entertainment, and surveillance in smart home environments. In this paper, an innovative HAR system, exploiting the potential of wearable devices integrated with the skills of deep learning techniques, is presented with the aim of recognizing the most common daily activities of a person at home. The designed wearable sensor embeds an inertial measurement unit (IMU) and a Wi-Fi section to send data on a cloud service and to allow direct connection to the Internet through a common home router so that the user themselves could manage the installation procedure. The sensor is coupled to a convolutional neural network (CNN) network designed to make inferences with the minimum possible resources to keep open the way of its implementation on low-cost or embedded devices. The system is conceived for daily activity monitor and nine different activities can be highlighted with an accuracy of 97%.

IoT Wearable Sensor and Deep Learning: An Integrated Approach for Personalized Human Activity Recognition in a Smart Home Environment

In this paper, we propose a method of human activity recognition with high throughput from raw accelerometer data applying a deep recurrent neural network (DRNN), and investigate various architectures and its combination to find the best parameter values. The “high throughput” refers to short time at a time of recognition. We investigated various parameters and architectures of the DRNN by using the training dataset of 432 trials with 6 activity classes from 7 people. The maximum recognition rate was 95.42% and 83.43% against the test data of 108 segmented trials each of which has single activity class and 18 multiple sequential trials, respectively. Here, the maximum recognition rates by traditional methods were 71.65% and 54.97% for each. In addition, the efficiency of the found parameters was evaluated using additional dataset. Further, as for throughput of the recognition per unit time, the constructed DRNN was requiring only 1.347 ms, while the best traditional method required 11.031 ms which includes 11.027 ms for feature calculation. These advantages are caused by the compact and small architecture of the constructed real time oriented DRNN.

Deep recurrent neural network for mobile human activity recognition with high throughput

In this paper, we propose a method of human activity recognition with high throughput from raw accelerometer data applying a deep recurrent neural network (DRNN), and investigate various architectures and its combination to find the best parameter values. The "high throughput" refers to short time at a time of recognition. We investigated various parameters and architectures of the DRNN by using the training dataset of 432 trials with 6 activity classes from 7 people. The maximum recognition rate was 95.42% and 83.43% against the test data of 108 segmented trials each of which has single activity class and 18 multiple sequential trials, respectively. Here, the maximum recognition rates by traditional methods were 71.65% and 54.97% for each. In addition, the efficiency of the found parameters was evaluated by using additional dataset. Further, as for throughput of the recognition per unit time, the constructed DRNN was requiring only 1.347 [ms], while the best traditional method required 11.031 [ms] which includes 11.027 [ms] for feature calculation. These advantages are caused by the compact and small architecture of the constructed real time oriented DRNN.

Deep Recurrent Neural Network for Mobile Human Activity Recognition with High Throughput

Path planning is an important function for executing autonomous moving robots, and many path planning methods that satisfy various constraints, such as avoiding obstacles and energy efficiency, have been proposed. However, these conventional methods have several difficulties for apply to the actual applications, such as the instability, low reproducibility, huge training data set required. Therefore, we propose a novel robot path planning method that combines the rapidly exploring random tree (RRT) and long short-term memory (LSTM) network. In this method, numerous and good paths are generated in the robot configuration space by the RRT method, a convolutional autoencoder and LSTM combination network is trained by them. The proposed method overcomes the difficulty of general methods with neural networks, i.e., “the acquisition of a large amount of training data.” Moreover, the difficulty of general random based methods, i.e., “the reproducible path generation” is resolved with high-speed.

Robot Path Planning by LSTM Network Under Changing Environment

We propose a robot path planning method adaptable to environmental change combining RRT and LSTM network. In this method, assuming multiple environments, a large amount of routes are generated by the RRT method and learning is performed using the LSTM network. We also try to adapt to environmental changes by using CAE during learning. By the proposed method, we perform the difficulty of a general random base method, that is, “generate reproducible route” at high speed. In addition, it is possible to generate routes adapted to small environmental changes.

Masaya Inoue

Papers

Deep recurrent neural network for mobile human activity recognition with high throughput

Deep Recurrent Neural Network for Mobile Human Activity Recognition with High Throughput

Robot Path Planning by LSTM Network Under Changing Environment

Recurrent Neural Network and Rapidly-exploring Random Tree Path Planning Adaptable to Environmental Change

Anomaly Detection for Solar Panel Joint Clamps Using Deep Learning Based Detection and Recognition Techniques