Abstract

Users navigating the web often deviate from the intended path and find themselves in a state of being lost. Early research shows, different cognitive factors such as domain knowledge, spatial ability, working memory, and prior experience contributes to the successful navigation. Existing cognitive models such as CoLiDeS, CoLiDeS+, CoLiDeS+Pic, MESA, and SNIF-ACT provide deeper analysis to understand the user behavior on the web. At the end, the goal of these models is to help the practitioners of the web in evaluating “the effectiveness of their designs”, in terms of user navigation. These models consider the selection process of a link as the key step in navigation and do so by measuring the semantic similarity of the link with respect to the goal. However, they do not completely account for all types of cognitive factors, and also the position of the links and content on a web page in the selection process. In this work, we present the results from two eye tracking studies conducted to understand the user behaviour during the link selection process. We observed that the users select a link with maximum similarity to the goal (information scent) and the information scent is always increased in a successful navigation path. We have also observed that the position and the information scent has not affected the way users attend the links. Although existing models do not consider the content while evaluating the links, we noticed that users spent significant amount of time in attending the content during the selection process. By comparing CoLiDes+ with Information Search process, we also found that the users feel more uncertain during the link selection stage of CoLiDes+. Incorporating all these observations, we modelled a tool to assist in navigation to reduce the user uncertainty and disorientation. While the above studies consider the user behavior as goal-oriented, often users browse without any predefined goal to start with. In this context, we carried out a behavioral study to understand the role of cognitive factors spatial ability and prior knowledge, on information seeking patterns. We observed the high spatial ability resulted in surface navigation whereas the low spatial ability resulted in deep navigation pattern. Users with high domain knowledge spent more time per page and visited less number of pages. These findings lead us to design an assistive feature by presenting the information from other sections on a given page. We believe this will result in better information discovery for the low spatial ability and the domain knowledge.

Abstract

Handwriting recognition (HWR) in Indic scripts is a challenging problem due to the inherent subtleties in the scripts, cursive nature of the handwriting and similar shape of the characters. Though a lot of research has been done in the field of text recognition, the focus of the vision community has been primarily on English. Furthermore, a lack of publicly available handwriting datasets in Indic scripts has also affected the development of handwritten word recognizers and made direct comparisons across different methods an impossible task in the field. Also, due to this lack annotated data, it becomes challenging to train deep neural networks which contain millions of parameters. These facts are quite surprising considering the fact that there are over 400 million Devanagari speakers in India alone. We first tackle the problem of lack of annotated data using various approaches. We describe a framework for annotating large scale of handwritten word images without the need for manual segmentation and label re-alignment. Two new word level handwritten datasets for Telugu and Devanagari are released which were created using the above mentioned framework. We synthesize synthetic datasets containing millions of realistic images with a large vocabulary for the purpose of pre-training using publicly available Unicode fonts. Later on, pre-training using data from Latin script is also shown to be useful to overcome the shortage of data.Capitalizing on the success of the CNN-RNN Hybrid architecture, we propose various improvements in the architecture and it’s training pipeline to make it even more robust for the purposes of handwriting recognition. We now change the network to use a Resnet-18 like structure for the convolutional part along with adding a spatial transformer network layer. We also use an extensive data augmentation scheme involving multi-scale, elastic, affine and test time distortion. We outperform the previous state-of-the-art methods on existing benchmark datasets for both Latin and Indic scripts by quite some margin. We perform an ablation study to empirically show how the various changes we made to the original CNN-RNN Hybrid network have improved its performance with respect to handwriting recognition. We dive deeper into the working of our networks convolutional layers and verify the robustness of convolutional-features through layer visualizations. We hope the release of the two datasets mentioned in this work along with the architecture and training techniques that we have used instill interest among fellow researchers of the field.

Abstract

Videos engulf the media archive in every capacity, generating a rich and vast source of information from which machines can learn how to understand such data and predict useful attributes that can help technologies make human lives better. One of the most significant and instrumental part of a computer vision system is the comprehension of human actions from visual sequences – a paragon of machine intelligence that can be achieved through computer vision. The problem of human action recognition has high importance as it facilitates several applications built around recognition of human actions. Understanding actions from monocular sequences has been studied immensely, while comprehending human actions from a skeleton video has developed in recent times. We propose action recognition frameworks that use the skeletal data (viz. 3D locations of some joints) of human body to learn spatio-temporal representations necessary for recognizing actions. Information about human actions is composed of two integral dimensions, space and time, along which the variations occur. Spatial representation of a human action aims at encapsulating the config- uration of human body essential to the action, while the temporal representation aims at capturing the evolution of such configurations across time. To this end, we propose to use geometric relations betweenhuman skeleton joints to discern the body pose relative to the action and physically inspired kinematic quantities in order to understand the temporal evolution of body pose. Spatio-temporal understanding of human actions is thus conceived as a comprehension of geometric and kinematic information with the help of machine learning frameworks. Using a representation inculcating an amalgamation of geo- metric and kinematic features, we recognize human actions from skeleton videos (S-videos) using such frameworks. We first present a non-parametric approach for temporal sub-segmentation of trimmed action videos using angular momentum trajectory of the skeletal pose sequence. Meaningful summarization of the pose sequence is a product of the sub-segmentation achieved through systematic sampling of the seg- ments. Descriptors capturing geometric and kinematic statistics encoded as histograms and spread across a periodic range of orientations are computed to represent the summarized pose sequences, which are fed to a kernelized classifier for recognizing actions. This framework for understanding human ac- tions instils the effects of using geometric and kinematic properties of human pose sequences, important to spatio-temporal modeling of the actions. However, a downside of this framework is the inability to scale with availability of large amount of visual data.To mitigate the impending drawback, we next present geometric deep learning frameworks and specifically, graph convolutional networks, for the same task. Representation of human skeleton as a sparse spatial graph is intuitive and a structured form which lies on graph manifolds. A human action video hence results in the formation of a spatio-temporal graph and graph convolutions facilitate the learning of a spatio-temporal descriptor for the action. Inspired by the success of Deformable Part-based Models (DPMs) for the task of object understanding from images, we propose a part-based graph con- volutional network (PB-GCN) that operates on a human skeletal graph divided into parts. Incorporating the culmination of understandings from the success of geometric and kinematic features, we propose to use relative coordinates and temporal displacements of the 3D joints coordinates as features at the vertices in the skeletal graph. Owing to these signals, the prowess of graph convolutional networks is further boosted to attain state-of-the-art performance among several action recognition systems using skeleton videos. In this thesis, we meticulously examine the growth of the idea about using geometry and kinematics, transition to geometric deep learning frameworks and design a PB-GCN with geometric + kinematic signals at the vertices, for the task of human action recognition using skeleton videos.

Abstract

360° field of view (FOV) is becoming a vital part of various human-robot/computer applications, such as urban search and rescue task. Despite growing importance of 360° FOV, very few studies have assessed the effect of 360° visual display on users' remote spatial ability, specifically in the case of 2D 360° User Interface (UI), which is the focus of the current study. We aim to evaluate the efficacy and efficiency of desktop 360° display designs on teleoperators' egocentric spatial ability, specifically navigation and direction judgment in an unknown environment. A previous study[1] showed an advantage of the segmented interface over seamless 360° UI on angle estimation error. However, we are still not clear as to the determining factors that lead to a difference in interface interaction ability. We conducted two studies that studied the scanning effort while using a 360° interface and the ability to estimate the mental egocentric bearing of a target in an environment. We measured the effects of gaming experience, and time constriction, with metrics consisting of time to task completion, target bearing estimation, and latency of target estimation. Further data was collected by an eye-tracking device to help support and complement the data from the studies. We found that gaming experience is a significant positive factor in accuracy and time to task completion. Further, a seamless panoramic view was found to be least effective compared to a segmented interface view of the 360° video feed for target bearing estimation across the entire field of vision presented through the interfaces. Eye tracking data supports the behavioral data collected. It was found that the estimation frequencies in the first study were high for forward heading direction, which was seen in the increased eye movement activity in the same region. In the second study, it was found that angle estimation error was the lowest along mental egocentric axes. This was again observed in the eye tracker data that reported increased activity along the mental egocentric axes.

Abstract

Forest fire is a major ecological disaster, which has economic, social and environmental impacts on humans and also causes the loss of biodiversity. Forest officials issue warnings to the public on the basis of fire danger index classes. Generally, fire danger indices are developed based on the meteorological stations in countries like Canada, United States of America, Australia, etc. Geospatial techniques such as satellite remote sensing based approaches can be useful to develop the fire danger indices in those countries that lack sufficient meteorological stations. In general, fire danger indices have been developed based on the parameters which are associated for the cause of ignition and spreading of forest fires. These properties include forest fuel type, topographic conditions and moisture conditions. Vegetation and topographic conditions are static, i.e. they do not change frequently, whereas moisture conditions are dynamic. Dynamic properties such as air temperature, relative humidity, moisture conditions changes regularly in a day. In this study, Static Fire danger Index has been developed using MODIS (Moderate Resolution Imaging Spectro Radiometer) Land cover type yearly L3 global 500 m SIN grid (MCD12Q1) and ASTER Global Digital Elevation Model datasets. International Geosphere-Biosphere Programme (IGBP) land cover type has been generated from MCD12Q1, which has been used to compute the forest fuel type index based on historical fire data. Fuel type danger index, Terrain ruggedness danger index, Slope danger index, Aspect danger index and Elevation danger index were computed from the ASTER GDEM datasets. Uttarakhand state has very few meteorological stations so geospatial techniques can be useful to derive the fire danger indices. Several authors developed the fire danger indices based on satellite derived parameters such as land surface temperature, vegetation and moisture indices such as “Normalized Difference Vegetation Index, Normalized Difference Water Index, Normalized Multiband Drought Index, Visible Atmospheric Resistant Index” etc. In this study, Dynamic Fire Danger Index (DFDI) has been developed from three parameters i.e. potential surface temperature, Perpendicular Moisture Index (PMI) and Modified Normalized Multiband Fire Index (MNDFI) using the MODIS Terra satellite datasets. DFDI has been calculated from the Near Real Time (NRT) Level 2 MODIS Terra Land Surface Temperature datasets (MOD11_L2) and MODIS Terra NRT surface reflectance dataset (MOD09). Finally, Forest Fire Danger Index (FFDI) has been developed by integrating both the Static and Dynamic fire danger indices and also used the near real time data sets that can be available for download through NASA FTP server Developing Forest Fire Danger index using geo-spatial techniques after one hour of the satellite overpass. The overall fire danger prediction accuracy was around 81.27% for the year 2016. Thus, the FFDI has been useful to assess the fire danger accurately over the study area and can be useful anywhere, where the meteorological stations are un-available. The entire procedure of calculating FFDI from NRT datasets was semi-automated so that the fire danger maps will be disseminated to the fire officials for taking timely action in controlling the forest fires.

Abstract

Sentiment analysis is a discipline of Natural Language Processing which deals with identifying and analyzing the subjectivity of data. It is an important task with both commercial and academic function- ality. It has acquired wide commercial uses including social media monitoring tasks, survey responses, review systems, etc. In review systems and surveys, sentiment analysis allows us to understand the opinions of multiple recipients. It expresses opinions such as like/dislike, for/against of general population. The merits of sentiment analysis do not stop there. The applications of sentiment analysis are broad and powerful. The ability to extract meaningful insights from social data is being widely adopted by organizations across the world.The human language is complex, teaching a machine to analyze the various grammatical nuances, cultural variations, slang and misspellings that occur in online mentions is a difficult process. To perform various sentiment analysis tasks we created a platform to employ different techniques and statistical methods to evaluate the data. By doing text analysis using computational linguistic techniques, we are able to capture and quantify affective states and subjective information. Sentiment analysis/opinion mining is done on human generated data. Since sentiment knowledge in humans grow with age and daily interactions, cognitive knowledge is required to understand emotion. In order for machines to understand underlying emotion, we need to transfer this cognitive knowledge in some form. Majority of the computational approaches to opinion mining comes using pre-existing lexicons. It has been understood that such lexicons are vital for any sentiment analysis system. These lexicons are the closest representation of our cognitive knowledge. These computational lexicon wrap sentiment information to assist in the task of opinion mining. Most research efforts in sentiment lexicon creation are bound to the English language. SentiWordNet for English is one such important lexical resource that contains subjective polarity for each lexical item. It was devised to aid in the task of opinion mining and sentiment classification. With growing data in native vernacular, there is a need for language-specific sentiment lexicon. For resource poor languages, creating such a sentiment lexicon is a difficult task to achieve. Described in this thesis are two different methods on how to build such a resource for an Indian language Tamil, using existing resources in other languages. Such a resource would serve as a baseline for further improvements in the task of sentiment analysis specific to Tamil data. Furthermore, the architecture followed in the two methods can be easily adapted to other languages.

Abstract

Forests play a key role in carbon cycle by sequestering the atmospheric carbon dioxide from various sources. Approximately 33% of the world’s land area is covered by forest ecosystems. It is thus essential to monitor forests at a regular interval for better management. Remote sensing datasets (both satellite and aerial) forms an key sources for providing a synoptic view of forest cover at landscape level to describe land use changes for monitoring purposes. However, the complexities of the forest ecosytems are not well understood with the traditional 2D remote sensing datasets (optical and radar remote sensing images) and need for the 3D structural assessment of forest ecosystems are essential. Terrestrial Laser Scanning (TLS) or Terrestrial LiDAR (Light Detection And Ranging), is an active remote sensing based ground measurement technique capable of capturing 3D information of the scene of interest with milli-meter accuracy. It transmits the laser pulse and analyses the return pulse to position the reflected object in 3D space. The main goal of this thesis is to evaluate the potential of TLS for forest biomass estimation and to develop the relevant methods to extract relevant information from TLS data. Accurate estimations of forest aboveground biomass at regional and national scales heavily depend on the field data and are upscaled using remote sensing datasets (both optical and radar). However,Field data collection of forest inventory parameters (mainly species name, tree count, tree diameter, height etc.,) using conventional methods are both labour-intensive and time consuming. So, firstly, we target the feasibility of TLS to extract forest inventory parameters accurately using the automated algorithm. We provide an automated solution to extract forest inventory parameters at individual tree level from 3D TLS data. The algorithm is tested over deciduous forests of Betul, Madhya Pradesh. Data filtering followed by a RANSAC based circle fitting algorithm was used to extract tree diameter and tree heights. An overall detection rate of 97% along with high correlation of field based easurements and TLS based measurements (Diameter R2 = 0.99 & Height R2 = 0.98) was obtained. The estimated RMSE was found to be 2.2 cm in diameter and 1.65 m in height. Further, to envisage the potential of TLS in diverse forestry applications, it was desirable to identify individual trees in the 3D TLS data to better describe tree architecture. So the focus was made to segment individual trees from the 3D point cloud by employing supervoxels and graph-cut based segmentation methods. High dense point cloud was first converted into supervoxels to reduce the data complexity and a graph-cut was performed to extract individual tree objects. The individual trees paved way for the advanced inventory parameter extraction such as crown dimensions (diameter and height) and branching patterns which would have been near impossible using conventional field measurements. Also, they form as the basic ground information for high resolution image based studies focusing on crown detection and delineations. Next, trees with relatively high completeness are selected and are described by highly accurate piecewise cylinders based on branching structure. This method helped to estimate the tree volume as the sum of the volumes of individual cylinders. Thus, an attempt has been made to accurately describe the actual tree volume in a non-destructive way. The volumes derived for selected trees were used to construct local volume equations of the study site along with the uncertainty in the total volume estimate. Tree volumes are a direct proxy of tree biomass (biomass is generally calculated by multiplying tree volumes with wood density), which help in carbon stock assessment of the forest area. Based on our estimates, tree volumes derived from conventional volume equations have underestimated the actual TLS volume by 28%. However, this estimate needs to be substantiated with destructive sampling. Also, the volume equations constructed using TLS along with uncertainty estimates are used in upscaling the field measurements to the regional biomass estimates using satellite remote sensing datasets. The standard way of producing a regional biomass map is by using a regression model using the field biomass estimates with the remote sensing variable (optical or radar). However, the uncertainty of this measurement at the spatial scale is often not calculated accurately. Both field level and the remote sensing model level errors attribute to the uncertainty of the spatial biomass estimate. Accurate assessment of biomass is essential to several social and economic incentive schemes of United Nations Framework Convention on Climate Change (UNFCCC). In this context, a possible framework to integrate uncertainties from field level measurements to regional-level estimates was established and spatial biomass maps along with uncertainty maps are gene

Abstract

Sentiment analysis is a discipline of Natural Language Processing which deals with identifying and analyzing the subjectivity of data. It is an important task with both commercial and academic function- ality. It has acquired wide commercial uses including social media monitoring tasks, survey responses, review systems, etc. In review systems and surveys, sentiment analysis allows us to understand the opinions of multiple recipients. It expresses opinions such as like/dislike, for/against of general population.The merits of sentiment analysis do not stop there. The applications of sentiment analysis are broad and powerful. The ability to extract meaningful insights from social data is being widely adopted by organizations across the world. The human language is complex, teaching a machine to analyze the various grammatical nuances,cultural variations, slang and misspellings that occur in online mentions is a difficult process. To perform various sentiment analysis tasks we created a platform to employ different techniques and statistical methods to evaluate the data. By doing text analysis using computational linguistic techniques, we are able to capture and quantify affective states and subjective information. Sentiment analysis/opinion mining is done on human generated data. Since sentiment knowledge in humans grow with age and daily interactions, cognitive knowledge is required to understand emotion. In order for machines to understand underlying emotion, we need to transfer this cognitive knowledge in some form. Majority of the computational approaches to opinion mining comes using pre-existing lexicons. It has been understood that such lexicons are vital for any sentiment analysis system. These lexicons are the closest representation of our cognitive knowledge. These computational lexicon wrap sentiment information to assist in the task of opinion mining. Most research efforts in sentiment lexicon creation are bound to the English language. SentiWordNet for English is one such important lexical resource that contains subjective polarity for each lexical item. It was devised to aid in the task of opinion mining and sentiment classification. With growing data in native vernacular, there is a need for language-specific sentiment lexicon. For resource poor languages, creating such a sentiment lexicon is a difficult task to achieve. Described in this thesis are two different methods on how to build such a resource for an Indian language - Tamil, using existing resources in other languages. Such a resource would serve as a baseline for further improvements in the task of sentiment analysis specific to Tamil data. Furthermore, the architecture followed in the two methods can be easily adapted to other languages.

Abstract

Gone are the days when individuals aimlessly pursued news from one source. Presently, they need to peruse articles and stories from an assortment of sources to see every one of its subtleties. News aggregators become an integral factor here. They gather news articles from an assortment of sources and present it to the client in one single area. With news originating from different streams, it turns out to be exceptionally cumbersome for a client to choose articles of her decision from a rundown of exhibited articles which relates to an assortment of subjects. In this thesis, we mainly focus on three major aspects of Computational Journalism which could help news aggregators to increase viewership as well as their credibility. We start with news recommendation, which is also one of the most crucial things for a news aggregator. Users, these days don’t have time to browse through hundreds of news articles and find the relevant ones. They want a recommendation system which could take into account their reading history and recommend relevant articles. We then look at a new problem which is used by news agencies to lure users into clicking at headlines and subsequently increase their viewership, which is formally known as Clickbait. Finally, we try to detect Fake News which could help news agencies in building trust and increase their credibility. We commence our research work with the problem of News Recommendation. Popular methods like collaborative filtering and content-based filtering have their own disadvantages. The former method requires a considerable amount of user data before making predictions, while the latter, suffers from over-specialization. In this work, we address both of these issues by coming up with a hybrid approach based on neural networks for news recommendation. The hybrid approach incorporates for both (1) user-item interaction and (2) content-information of the articles read by the user in the past. We first come up with an article-embedding based profile for the user. We then use this user profile with adequate positive and negative samples in order to train the neural network-based model. We then move on to various deep learning methods to tackle the same problem. We propose a Re-current Attention Recommendation Engine (RARE) which consists of two components and utilizes the distributed representations of news articles. The first component is used to model the user’s sequential behaviour of news reading in order to understand her general interests, i.e., to get a summary of her interests. The second component utilizes an article level attention mechanism to understand her specific interests. We feed the information obtained from both the components to a Siamese Network in order to make predictions which pertain to the user’s generic as well as specific interests. Building upon our previous model, we propose a Hybrid Recurrent Attention Machine (HRAM).HRAM consists of two components. The first component utilizes a neural network for matrix factoriza-tion. While in the second component, we first learn the distributed representation of each news article.We then use the historical data of the user in a sequential manner and feed it to an attention-based recurrent layer. Finally, we concatenate the outputs from both these components and use it to make predictions. In this way, we harness the information present in the user reading history and boost it with the information available through collaborative filtering for providing better news recommendations. We then experiment with non-recurrent neural networks for the same problem. We propose a simple yet effective architecture which utilizes a 3D Convolutional Neural Network which takes the word em-beddings of the articles present in the user history as its input. Using such a method endows the model with the capability to automatically learn spatial (features of a particular article) as well as temporal features (features across articles read by a user) which signify the interest of the user. At test time, we use this in combination with a 2D Convolutional Neural Network for recommending articles to users. We carry out extensive experiments to establish the effectiveness of our methods. We also perform experiments to prove the efficacy of our model on solving the item cold-start cases as well as making effective recommendations for users who have had very little interaction with articles. We then look at another aspect of Computational Journalism, which is Clickbait Detection. Online media outlets, in a bid to expand their reach and subsequently increase revenue, have begun adopting clickbait techniques to lure readers to click on articles. We propose a novel approach considering all information found in a social media post. We train a bidirectional LSTM with an attention mechanism to learn the extent to which a word contributes to the post’s clickbait score in a differential manner. We also employ a Siamese net

Abstract

The problem of developing a good humanoid platform that can operate in diverse environments is very demanding as it not only requires a considerable amount of time in hardware development but also takes into account significant effort in software development. It requires to go through various iterations to zero down on the design and remove any other potential issues associated with the hardware. In this thesis, we focus on design and development of one such humanoid platform which is modular, open source, stable, cost-effective and easy to build using modern techniques like rapid prototyping. The first and foremost reason for the development of such a platform is that it can be very useful in verification of the various algorithms, which are critical for the operation of humanoid. Secondly, in a long run, it may help in understanding the complexities of human motion in various environments and then replicate it efficiently on the robot. Such an evolution may help us to revolutionize Human-Robot Interaction (HRI). Such a humanoid robot can evolve and learn, allowing for safer interaction without any harm to either robot or the user. But there are inherent problems associated with this notion of development of software and hardware for humanoids. For humans, walking and balancing seem like a trivial task which comes naturally to us, but such is not the case for the robots. They have to be programmed and equipped with sufficient feedback to leverage their body structure and then stand in a stable posture accordingly with continuous control and feedback. A very prominent factor in achieving this stability is that humanoids have to be morphologically designed which assist in the manipulation tasks or gait to be performed by the robot. The robot should have sufficient motor torque and longevity to stand and should be able to take into account the dynamic forces acting on the body. Motors should be able to execute the given motion profiles concurrently so that the robot doesn’t fall due to conflict of joint motion. The electrical hardware has to be powerful enough to have good response time and reliability. Since there are a lot of variations in sizes of humanoids and difficulties associated with each type and sizes, It was decided to go for a miniature humanoid as it is relatively easier and cheaper to build and it still allows exploration of most important research problems. Building on Poppy project, INRIA,France, we develop a humanoid with a 5-Degree-of-Freedom (5-DoF) torso with some key changes. The benefit of such a torso is enormous as it allows for distribution of impact forces evenly on the body without destabilizing it. In contrast to Poppy robot, the ankle joint of the leg has 2-DoF joint in place of the 1-DoF joint. This enables the robot to adapt to different terrains more effectively. Humanoid was made using the rapid prototyping techniques, which reduces the hardware construction time and any reiteration occurring due to flaws. The major efforts have been made in building a humanoid platform with robust hardware and feedback modules. This is achieved by designing the structure of the robot, ensuring proper joint motion feedback, and enabling feedback from the environment through sensors in a reliable and robust manner. All these problems require painstaking effort to build and test for the whole system to work in a holistic manner. We have tried to keep the structure of the robot as close to a human morphology as possible and showed that it is stable when it stands and can perform necessary motor manipulations with large enough support polygon formed by the extremities of the foot. Incorporation of sensors like force sensing resistors, position sensors, IMU, etc. with an embedded hardware helped in coming up with a reliable platform. This was followed by the development of the algorithm for joint control by commanding the individual motors of the humanoid concurrently. The stability of humanoid is monitored by experimental calculation of Zero Moment Point (ZMP) which enables future usage of the hardware to test complicated motion planning algorithms. In order to work safely with the hardware, a MATLAB/Simulink based simulator is also developed which allows the first level of testing through simulations. This helps in understanding the implications of the developed algorithms before implementing on the real robot. With the simulation environment in place, It is easy to directly connect such an environment and the robot using a middle-ware like ROS to perform experimental validation. The platform is intended to be used for testing of various algorithms of humanoid motion in various