Abstract: Simulation of dynamic physical systems is done using iterative solvers. However, this iterative process is a time consuming and compute intensive process and, for a given set of simulation parameters, the solution does not always converge to a physically meaningful solution, resulting in huge waste of man hours and computation resource. Embodiments herein provide a method and system for stabilizing a diverged numerical simulation and accelerating a converged numerical simulation by changing one or more control parameters. An automatic monitoring mechanism of residue history (to interpret convergence or divergence) and a subsequent control logic to auto-tune the under-relaxation factor would help in stabilizing a diverging simulation and reaching faster convergence by accelerating converging simulation.

Abstract: Conventional gesture detection approaches demand large memory and computation power to run efficiently, thus limiting their use in power and memory constrained edge devices. Present application/disclosure provides a Spiking Neural Network based system which is a robust low power edge compatible ultrasound-based gesture detection system. The system uses a plurality of speakers and microphones that mimics a Multi Input Multi Output (MIMO) setup thus providing requisite diversity to effectively address fading. The system also makes use of distinctive Channel Impulse Response (CIR) estimated by imposing sparsity prior for robust gesture detection. A multi-layer Convolutional Neural Network (CNN) has been trained on these distinctive CIR images and the trained CNN model is converted into an equivalent Spiking Neural Network (SNN) via an ANN (Artificial Neural Network)-to-SNN conversion mechanism. The SNN is further configured to detect/classify gestures performed by user(s).

Abstract: The present disclosure relates to methods and systems for time-series classification using a reservoir-based spiking neural network, that can be used at edge computing applications. Conventional reservoir based SNN techniques addressed either by using non-bio-plausible backpropagation-based mechanisms, or by optimizing the network weight parameters. The present disclosure solves the technical problems of TSC, using a reservoir-based spiking neural network. According to the present disclosure, the time-series data is encoded first using a spiking encoder. Then the spiking reservoir is used to extract the spatio-temporal features for the time-series data. Lastly, the extracted spatio-temporal features of the time-series data is used to train a classifier to obtain the time-series classification model that is used to classify the time-series data in real-time, received from edge devices present at the edge computing network.

Abstract: Conventional gesture detection approaches demand large memory and computation power to run efficiently, thus limiting their use in power and memory constrained edge devices. Present application/disclosure provides a Spiking Neural Network based system which is a robust low power edge compatible ultrasound-based gesture detection system. The system uses a plurality of speakers and microphones that mimics a Multi Input Multi Output (MIMO) setup thus providing requisite diversity to effectively address fading. The system also makes use of distinctive Channel Impulse Response (CIR) estimated by imposing sparsity prior for robust gesture detection. A multi-layer Convolutional Neural Network (CNN) has been trained on these distinctive CIR images and the trained CNN model is converted into an equivalent Spiking Neural Network (SNN) via an ANN (Artificial Neural Network)-to-SNN conversion mechanism. The SNN is further configured to detect/classify gestures performed by user(s).

Abstract: Disadvantage of these existing approaches for spike encoding optimization are that they fail to process multivariate data and perform energy-efficient time series classification at edge. The disclosure herein generally relates to spike encoding optimization, and, more particularly, to a method and system for Mutual Information (MI) based spike encoding optimization. Before performing spike data optimization for a given input multivariate time series data, the system ensures by iteratively adding gaussian noise to the input data that the input data has achieved a maximum MI value. After the input data has achieved a maximum MI value, spike train optimization is done by the system, to generate optimized spike data and in turn a spike reservoir.

Abstract: This disclosure relates generally to time series forecasting, and, more particularly, to a system and method for online time series forecasting using spiking reservoir. Existing systems do not cater for efficient online time-series analysis and forecasting due to their memory and computation power requirements. System and method of the present disclosure convert a time series value F(t) at time ‘t’ to an encoded multivariate spike train and extracts temporal features from the encoded multivariate spike train by the excitatory neurons of a reservoir, predict a time series value Y(t + k) at time ‘t’ by performing a linear combination of extracted temporal features with read-out weights, compute an error for predicted time series value Y(t + k) with input time series value F(t + k), employs a FORCE learning on read-out weights using the error to reduce error in future forecasting. Feeding a feedback value back to the reservoir to optimize memory of the reservoir.

Abstract: Gesture recognition is a key requirement for Human Computer Interaction (HCI) and multiple modalities are explored in literature. Conventionally, channel taps are estimated using least square based estimation and tap corresponding to finger motion is tracked. These assume that noise component is negligible and can reduce the tracking accuracy for low SNR. Thus, to mitigate the above-mentioned limitation, the system and method of the present disclosure explore the feasibility of using speaker and microphone setup available in most of smart devices and transmit inaudible frequencies (acoustic) for detecting the human finger level gestures accurately. More specifically, System implements the method for millimeter level finger tracking and low power gesture detection on this tracked gesture. The system uses a subspace based high resolution technique for delay estimation and use microphone pairs to jointly estimate the multi-coordinates of finger movement.

Abstract: State of art techniques rely of FPGA based approaches when power efficiency is of concern. However, compared to SNN on Neuromorphic hardware, ANN on FPGA requires higher power and longer design cycles to deploy neural network on hardware accelerators. Embodiments of the present disclosure provide a method and system for energy efficient hierarchical multi-stage SNN architecture for classification and segmentation of high-resolution images. Patch-to-patch-class classification approach is used, where the image is divided into smaller patches, and classified at first stage into multiple labels based on percentage coverage of a parameter of interest, for example, cloud coverage in satellite images. The image portion corresponding to the partially covered patches is divided into further smaller size patches, classified by a binary classifier at second level of classification.

Abstract: This disclosure relates generally to a method and a system for computing using a field programmable gate array (FPGA) neuromorphic architecture. Implementing energy efficient Artificial Intelligence (AI) applications at power constrained environment/devices is challenging due to huge energy consumption during both training and inferencing. The disclosure is a FPGA architecture based neuromorphic computing platform, the basic components include a plurality of neurons and memory. The FPGA neuromorphic architecture is parameterized, parallel and modular, thus enabling improved energy/inference and Latency-Throughput. Based on values of the plurality of features of the data set, the FPGA neuromorphic architecture is generated in a modular and parallel fashion. The output of the disclosed FPGA neuromorphic architecture is the plurality of output spikes from the neuron, which becomes the basis of inference for computing.

Abstract: This disclosure relates generally to robotic network, and more particularly to a method and system for hierarchical decomposition of tasks and task planning in a robotic network. While a centralized system is used for action planning in a robotic network, any communication network issues can adversely affect working of the robotic network. Further, hardcoding one or more specific tasks to a robot restricts use of the robots irrespective of capabilities of the robots. The robotic agent decomposes a goal assigned to the robot to multiple sub-goals, and for each sub-goal, identifies one or more tasks to be executed/performed by the robot. An action plan is generated based on all such tasks identified, and the robot executes the action plan, in response to the goal assigned to the robot.

Abstract: This disclosure generally relates optimized spike encoding for spiking neural networks (SNNs). The SNN processes data in spike train format, whereas the real world measurements/input signals are in analog (continuous or discrete) signal format; therefore, it is necessary to convert the input signal to a spike train format before feeding the input signal to the SNNs. One of the challenges during conversion of the input signal to the spike train format is to ensure retention of maximum information between the input signal to the spike train format. The disclosure reveals an optimized encoding method to convert the input signal to optimized spike train for spiking neural networks. The disclosed optimized encoding approach enables maximizing mutual information between the input signal and optimized spike train by introducing an optimal Gaussian noise that augments the entire input signal data.

Abstract: This disclosure relates to method of identifying a gesture from a plurality of gestures using a reservoir based convolutional spiking neural network. A two-dimensional spike streams is received from neuromorphic event camera as an input. The two-dimensional spike streams associated with at least one gestures from a plurality of gestures is preprocessed to obtain plurality of spike frames. The plurality of spike frames is processed by a multi layered convolutional spiking neural network to learn plurality of spatial features from the at least one gesture. A filter block is deactivated from the plurality of filter blocks corresponds to at least one gesture which are not currently being learnt. A spatio-temporal features is obtained by allowing the spike activations from CSNN layer to flow through the reservoir. The spatial feature is classified by classifier from the CSNN layer and the spatio-temporal features from the reservoir to obtain set of prioritized gestures.

Abstract: This disclosure relates to method of identifying a gesture from a plurality of gestures using a reservoir based convolutional spiking neural network. A two-dimensional spike streams is received from neuromorphic event camera as an input. The two-dimensional spike streams associated with at least one gestures from a plurality of gestures is preprocessed to obtain plurality of spike frames. The plurality of spike frames is processed by a multi layered convolutional spiking neural network to learn plurality of spatial features from the at least one gesture. A filter block is deactivated from the plurality of filter blocks corresponds to at least one gesture which are not currently being learnt. A spatio-temporal features is obtained by allowing the spike activations from CSNN layer to flow through the reservoir. The spatial feature is classified by classifier from the CSNN layer and the spatio-temporal features from the reservoir to obtain set of prioritized gestures.

Abstract: This disclosure relates generally to action recognition and more particularly to system and method for real-time radar-based action recognition. The classical machine learning techniques used for learning and inferring human actions from radar images are compute intensive, and require volumes of training data, making them unsuitable for deployment on network edge. The disclosed system utilizes neuromorphic computing and Spiking Neural Networks (SNN) to learn human actions from radar data captured by radar sensor(s). In an embodiment, the disclosed system includes a SNN model having a data pre-processing layer, Convolutional SNN layers and a Classifier layer. The preprocessing layer receives radar data including doppler frequencies reflected from the target and determines a binarized matrix. The CSNN layers extracts features (spatial and temporal) associated with the target's actions based on the binarized matrix.

Abstract: Systems and methods for generating control system solutions for robotics environments is provided. The traditional systems and methods provide robotics solutions but specialized to only a particular robotic application, domain, and selected structure.

Abstract: Parameters specific to robot, environment, target objects and their inter-relations need to be considered by a robot to estimate cost of a task. As the existing task allocation methods assume a single utility value for a robot-task pair, combining heterogeneous parameters is a challenge. In applications like search and rescue, manual intervention may not be possible in real time. For such cases, utility calculation may be a hindrance towards automation. Also, manufacturers follow their own nomenclature and units for robotic specifications. Only domain experts can identify semantically similar terms and perform necessary conversions. Systems and methods of the present disclosure provide a structured semantic knowledge model to store and describe data in a uniform machine readable format such that semantics of those data can be interpreted by the robots and utility computation can be autonomous to make task allocation autonomous, semantic enabled and capable of self-decision without human intervention.

Abstract: Path planning for a robot is a compute intensive task. For a dynamic environment this is more cumbersome where position and orientation of objects changes often. Embodiments of the present disclosure provide systems and methods for context based path planning for vector navigation in hexagonal spatial maps. A 2-D environment is represented into a hexagonal grid map that includes hexagonal grid cells, objects are identified based on a comparison of RGB value associated with contiguous cells. Candidate contexts are determined based on objects identified. The hexagonal grid map is rotated at various angles and compared with pre-defined map(s) to determine quantitative measure of similarity for contexts identification from the candidate contexts, based upon which a path is dynamically planned for easy and efficient vector navigation within the hexagonal grid map.

Abstract: This disclosure relates generally to robotics, and more particularly to collaborative task execution by a robotic group using a distributed semantic knowledge base. In one embodiment, a method and a system of collaborative task execution by a robotic group using a distributed semantic knowledge base are provided. The distributed semantic knowledge base is distributed between multiple robots that form a robotic group. Each robot triggers a self query as well as one or more external queries to gather required data pertaining to a plurality of task specific parameters, and uses the gathered data to execute one or more tasks assigned to the robotic group.

Abstract: Systems and methods for modelling prediction errors in path-learning of an autonomous learning agent are provided. The traditional systems and methods provide for machine learning techniques, wherein estimation of errors in prediction is reduced with an increase in the number of path-iterations of the autonomous learning agent. Embodiments of the present disclosure provide for a two-stage modelling technique to model the prediction errors in the path-learning of the autonomous learning agent, wherein the two-stage modelling technique comprises extracting a plurality of fitted error values corresponding to a plurality of predicted actions and actual actions by implementing an Autoregressive moving average (ARMA) technique on a set of prediction error values; and estimating, by implementing a linear regression technique on the plurality of fitted error values, a probable deviation of the autonomous learning agent from each of an actual action amongst a plurality of predicted and actual actions.

Abstract: Parameters specific to robot, environment, target objects and their inter-relations need to be considered by a robot to estimate cost of a task. As the existing task allocation methods assume a single utility value for a robot-task pair, combining heterogeneous parameters is a challenge. In applications like search and rescue, manual intervention may not be possible in real time. For such cases, utility calculation may be a hindrance towards automation. Also, manufacturers follow their own nomenclature and units for robotic specifications. Only domain experts can identify semantically similar terms and perform necessary conversions. Systems and methods of the present disclosure provide a structured semantic knowledge model to store and describe data in a uniform machine readable format such that semantics of those data can be interpreted by the robots and utility computation can be autonomous to make task allocation autonomous, semantic enabled and capable of self-decision without human intervention.