Hadi Moradi - Academia.edu (original) (raw)

Papers by Hadi Moradi

In this paper, the design and implementation of a novel passive safe joint is reported. This safe... more In this paper, the design and implementation of a novel passive safe joint is reported. This safe joint is designed for applications, such as an air hockey playing robot playing against a human, in which the force applied by the linkage should be kept below a certain threshold. The proposed mechanism benefits from the compliance of two ramps controlled by the nonlinear behavior of a spring. The passivity of the design, the small footprint, the low cost of production, and the ability to rotate 360 degrees are the main advantages of this design. The mechanism has been designed, evaluated, implemented and tested showing its effectiveness. Finally it has been incorporated in an air hockey robot showing its capability in action.

arXiv (Cornell University), Jun 1, 2018

Pose Graph Optimization (PGO) is an important non-convex optimization problem and is the state-of... more Pose Graph Optimization (PGO) is an important non-convex optimization problem and is the state-of-the-art formulation for SLAM in robotics. It also has applications like camera motion estimation, structure from motion and 3D reconstruction in machine vision. Recent researches have shown the importance of good initialization to bootstrap well-known iterative PGO solvers to converge to good solutions. The stateof-the-art initialization methods, however, works in low noise or eventually moderate noise problems, and they fail in challenging problems with high measurement noise. Consequently, iterative methods may get entangled in local minima in high noise scenarios. In this paper we present an initialization method 1 which uses orientation measurements and then present a convergence analysis of our iterative algorithm. We show how the algorithm converges to global optima in noise-free cases and also obtain a bound for the difference between our result and the optimum solution in scenarios with noisy measurements. We then present our second algorithm that uses both relative orientation and position measurements to obtain a more accurate least squares approximation of the problem that is again solved iteratively. In the convergence proof, a structural coefficient arises that has important influence on the basin of convergence. Interestingly, simulation results show that this coefficient also affects the performance of other solvers and so it can indicate the complexity of the problem. Experimental results show the excellent performance of the proposed initialization algorithm, specially in high noise scenarios.

In this paper a real-time approach for the air hockey puck's velocity estimation based on motion ... more In this paper a real-time approach for the air hockey puck's velocity estimation based on motion blur is presented. In the proposed approach first a low resolution image is used to detect the puck and then the Fourier transform is used on the full resolution image to extract the direction of the motion. This hierarchical approach provides faster processing speed. The roundness of the ball and the length of the blur are finally used in order to estimate the speed. The proposed approach is implemented on an actual air hockey table using two different cameras, one with 30 and the other one with 60 frames per second. The obtained results show promising performance on velocity estimation using off-the-shelf cameras.

International Journal of Social Robotics, Jul 5, 2015

Recent research shows that Human Robot Interaction (HRI) can play an important role in the screen... more Recent research shows that Human Robot Interaction (HRI) can play an important role in the screening and therapy of many disorders, and suggests broad range of applications for children with developmental disorders. This paper presents the design, development, and application of a parrot-like robot as a screening tool to diagnose autistic children. The interactions of several autistic children with the robot have been observed and compared to the interaction of a group of normal children with the robot to determine the important behavioral features separating the autistic children from the normal ones. Based on these observations, we developed a descriptive research methodology to specify important and discriminating features for screening autism using behavioral aspects in social interaction, communication, and stereotyped and repetitive behaviors. Our experiments show that there are significant differences between autistic and normal children pertaining to their communication with the robot, repetitive behaviors and sustaining attention. The results show the effectiveness of using robots in screening developmental disorders such as autism, highlighting the importance of Robot Assisted Screening (RAS) or intelligent toys in this field. We have determined 6 features extracted from the children’s behavior, and Random Forest method is used to distinguish between the autistic and normal children. The system has been tested on 51 children, 35 autistic and 16 normal, with maximum 90 % correct classification rate.

To study human motor control and learning in upper extremities, scientists widely use robotic arm... more To study human motor control and learning in upper extremities, scientists widely use robotic arms to apply external forces to subject's hand interactively. In this work, we describe the design procedure of a planar parallel manipulandum with five links and 2-DoF in a vertical plane (e.g. sagittal or frontal plane). The manipulandum works in two active and passive modes. In the passive mode, the robotic arm follows the movements of the human arm; while in the active mode, the robotic endeffector guides subject's hand in any desired direction. In both active and passive conditions, the robot arm can be programmed to apply unexpected disturbances to the subjects hand. To control the movement of the manipulandum, we used two PID controllers with local shaft angle feedback to control the manipulandum. For the active mode, we used also a global feedback from endeffector position. As the first step for proof of concept, the whole system was simulated in MATLAB.

arXiv (Cornell University), Mar 11, 2021

There are partially separable data types that make classification tasks very hard. In other words... more There are partially separable data types that make classification tasks very hard. In other words, only parts of the data are informative meaning that looking at the rest of the data would not give any distinguishable hint for classification. In this situation, the typical assumption of having the whole labeled data as an informative unit set for classification does not work. Consequently, typical classification methods with the mentioned assumption fail in such a situation. In this study, we propose a framework for the classification of partially separable data types that are not classifiable using typical methods. An algorithm based on the framework is proposed that tries to detect separable subgroups of the data using an iterative clustering approach. Then the detected subgroups are used in the classification process. The proposed approach was tested on a real dataset for autism screening and showed its capability by distinguishing children with autism from normal ones, while the other methods failed to do so.

arXiv (Cornell University), Nov 10, 2019

When the direct view between the target and the observer is not available, due to obstacles with ... more When the direct view between the target and the observer is not available, due to obstacles with non-zero sizes, the observation is received after reflection from a reflector, this is the indirect view or Non-Line-Of Sight condition. Localization of a target in NLOS condition still one of the open problems yet. In this paper, we address this problem by localizing the reflector and the target simultaneously using a single stationary receiver, and a determined number of beacons, in which their placements are also analyzed in an unknown map. The work is done in mirror space, when the receiver is a camera, and the reflector is a planar mirror. Furthermore, the distance from the observer to the target is estimated by size constancy concept, and the angle of coming signal is the same as the orientation of the camera, with respect to a global frame. The results show the validation of the proposed work and the simulation results are matched with the theoretical results.

International Journal of Robotics & Automation, 2016

The hybrid localization using Angle Of Arrival (AOA) and Di erential Received Strength Signal Ind... more The hybrid localization using Angle Of Arrival (AOA) and Di erential Received Strength Signal Indicator (DRSSI) of an RF source with unknown power and Non-Line-Of-Sight (NLOS) condition has been proven to be advantageous compared to using each method separately. In this paper, the initial hybrid method, which was implemented using particle lters due to the multi-modal/non-Gaussian nature of localization in NLOS condition, has been replaced by a multi-step Gaussian ltering approach which provides similar accuracy with better performance. This has been done using DRSSI input in the rst step of the ltering to determine the linearization point, and then using AOA and DRSSI inputs together in the second step of the ltering to improve the localization accuracy. The proposed method has been implemented using Extended Kalman lter and Unscented Kalman lter. The simulation results show that the accuracy of the multi-step Gaussian ltering is comparable to the particle ltering approach with much lower computational load that is important for online localization of several RF sources. Furthermore, the e ects of uncertainty on the propagation parameters have been studied to show that the robustness of the multi-step Gaussian ltering to the uncertainties is comparable to the particle lter approach.

IEEE Transactions on Robotics, Jun 1, 2015

ABSTRACT In a wheeled-tip manipulation system, the wheels at the tip of robot fingers grasp objec... more ABSTRACT In a wheeled-tip manipulation system, the wheels at the tip of robot fingers grasp objects and manipulate them by rolling over their boundaries. In this paper, we introduce a 2-D kinematic modeling for wheeled-tip manipulation systems in order to handle rolling over objects’ corners in the manipulation. The modeling facilitates motion planning through projecting grasp stability conditions into configuration space, which unifies the planning of the grasp stability and obstacle avoidance. It is done through defining a static force closure map independent of the obstacles and the robot mechanism. This results in a straightforward planning in PRM framework. The simulation and experimental results support the approach.

Recently video-based data collection has been widely used in intelligent transportation systems. ... more Recently video-based data collection has been widely used in intelligent transportation systems. For instance, traffic flow monitoring at an intersection is one of these systems used for transportation network analysis. Main problems in intersection monitoring are vehicle detection and tracking. In this paper, we have developed and implemented a video-based vehicle detection and tracking system. The vehicle detection method is based on a combination of background estimation and dynamic texture modeling. After extracting vehicles from the video frames, a point tracking method has been used for prediction of vehicles' central points in the next frames. Weighted recursive least square has been used for point tracking purpose. Moreover, to solve the occlusion of two or more vehicles problem, fast normalized cross correlation algorithm has been used as a template matching method. The reported experimental results, verify the effectiveness of proposed method in vehicle tracking when occlusions occurs.

Lecture Notes in Computer Science, 1998

The Robocup 97 competition provides an excellent opportunity to demonstrate the techniques and me... more The Robocup 97 competition provides an excellent opportunity to demonstrate the techniques and methods of artificial intelligence, autonomous agents and computer vision. On a soccer field the core capabilities a player must have are to navigate the field, track the ball and other agents, recognize the difference between agents, collaborate with other agents, and hit the ball in the correct direction. USC's Dreamteam of robots can be described as a group of mobile autonomous agents collaborating in a rapidly changing environment. The key characteristic of this team is that each soccer robot is an autonomous agent, self-contained with all of its essential capabilities on-board. Our robots share the same general architecture and basic hardware, but they have integrated abilities to play different roles (goalkeeper, defender or forward) and utilize different strategies in their behavior. Our philosophy in building these robots is to use the least possible sophistication to make them as robust as possible. In the 1997 RoboCup competition, the Dreamteam played well and won the world championship in the middlesized robot league.

Sustainability

Early screening, which can lead to early diagnosis and intervention for children with autism (ASD... more Early screening, which can lead to early diagnosis and intervention for children with autism (ASD), can significantly improve their life quality. Consequently, the observational process of the ASD diagnosis and the lack of experts make the technology-based ASD screening methods more demanding. On the other hand, early ASD screening based on behavioral patterns is one of the most reliable methods that could be accomplished by analyzing children’s playing patterns. Thus, in this paper, we present an extension of our initial sensorized toy car functionalities by adding shaft encoders to detect attention to details and interest in rotating objects in children with ASD. Using the two modalities to detect different ASD symptoms improved our screening accuracy by more than 10%. Having such a sensorized toy car, which can be used at homes or clinics for screening ASD without high expertise, can improve the life of children with ASD and their families by providing an initial screening for fu...

Frontiers in Robotics and AI

21st century brought along a considerable decrease in social interactions, due to the newly emerg... more 21st century brought along a considerable decrease in social interactions, due to the newly emerged lifestyle around the world, which became more noticeable recently of the COVID-19 pandemic. On the other hand, children with autism spectrum disorder have further complications regarding their social interactions with other humans. In this paper, a fully Robotic Social Environment (RSE), designed to simulate the needed social environment for children, especially those with autism is described. An RSE can be used to simulate many social situations, such as affective interpersonal interactions, in which observational learning can take place. In order to investigate the effectiveness of the proposed RSE, it has been tested on a group of children with autism, who had difficulties in emotion recognition, which in turn, can influence social interaction. An A-B-A single case study was designed to show how RSE can help children with autism recognize four basic facial expressions, i.e., happin...

Expert Systems with Applications

arXiv (Cornell University), Jan 28, 2021

Comments: 24 pages with 4 figures Category: cs Subj-class: Human Computer Interaction ACM-class: ... more Comments: 24 pages with 4 figures Category: cs Subj-class: Human Computer Interaction ACM-class: K.4.2 Robot Assisted Therapy (RAT) is a new paradigm in many therapies such as the therapy of children with autism spectrum disorder. In this paper, we present the use of a parrot-like robot as an assistive tool in turn-taking therapy. The therapy is designed in the form of a card game between a child with autism and his/her therapist or the robot. The intervention was implemented in a single subject study format and the effect sizes for different turn-taking variables are calculated. The results show that, the child-robot interaction had larger effect size than the childtrainer effect size in most of the turn-taking variables. Furthermore, the therapist's point of view on the proposed Robot Assisted Therapy is evaluated using a questionnaire. The therapist believes that the robot is appealing to children which may ease the therapy process. The therapist suggested to add other functionalities and games to let children with autism to learn more turn taking tasks and better generalize the learned tasks.

هلاقم تاعلاطا هدیکچ لماک یشهوژپ هلاقم :تفایرد 07 نمهب 1392 :شریذپ 18 نیدرورف 1393 :تیاس رد هئارا ... more هلاقم تاعلاطا هدیکچ لماک یشهوژپ هلاقم :تفایرد 07 نمهب 1392 :شریذپ 18 نیدرورف 1393 :تیاس رد هئارا 02 رهم 1393 نیا رد هلاقم تابر يارب بذاج متسیس عون کی یم فورعم یسکترو بذاج هب هک ،درون راوید ياه یهاگشیامزآ و يددع ،یلیلحت تروصب ،دشاب هظفحم لماش یسکترو بذاج متسیس .تسا هتفرگ رارق هعلاطم دروم یم كرحم روتوم و هناورپ ،بذاج ي اب ،بذاج نیا نورد سکترو هک دشاب زا هدافتسا (ژوفیرتناس نف) هناورپ یم دیلوت یم نآ نورد لایس نایرج ندوب روصحم ،بذاج متسیس نیا یگژیو نیرتزراب .دوش یحارط نیا هک دشاب یم روتوم یفرصم ناوت شهاک و هبذاج يورین شیازفا ثعاب تازیهجت هب زهجم بذاج متسیس کی ،ادتبا رد .دوش هزادنا و رود ،راشف نادیم يریگ هزادنا ،يدیلوت هبذاج يورین يور فلتخم ياهرتماراپ ریثات و هدش هتخاس يدورو ناوت اب هک دش هدهاشم شخب نیا رد .تسا هدش لیلحت و يریگ هلصاف شیازفا حطس زا بذاج ي ، یم شیازفا یفرصم ناوت .دبای هیبش مرن زا هدافتسا اب بذاج متسیس يددع يزاس رازفا CFX زا ،لح جیاتن و ماجنا تسار یهاگشیامزآ جیاتن اب هسیاقم و هکبش زا لح للاقتسا قیرط تسا هدش ییامزآ . یم ناشن هسیاقم نیا طاقن نیب توافت نیرتشیب هک دهد ربارب ابیر...

In this paper, RF source localization in non-line of sight condition, using map of the obstacles ... more In this paper, RF source localization in non-line of sight condition, using map of the obstacles is proposed. Received signal strength indicator (RSSI) and angle of arrival (AOA) measurements are observations which are obtained from received signal on the UAV. In the proposed approach, AOA are used to determine the obstacle on the map from which the reflection has happened. Then the RSSI information is used to determine the location of the RF source. In the basic version of approach, triangulation is used to determine the location of the RF source. In the advanced approach, the reflection angle is also estimated to improve the localization accuracy. The estimation is done using particle filter approach. In addition, it is shown analytically that the maximum localization error for the advanced approach is bounded but relative formation of the reflectors with respect to each other can increase the localization error for the basic approach.

Autism-Open Access, 2018

The main aim of the current study was to develop an expert system for screening autism spectrum d... more The main aim of the current study was to develop an expert system for screening autism spectrum disorders. The statistical properties of the system were further examined. To design an expert system, having a reliable and valid knowledge base (KL) is very important. To provide the knowledge base, items linked to autism diagnoses were collected from the literature and then reviewed by a group of psychologists and psychometrists experts. The questionnaire was completed by parents of children with autism (85), parents of normal children (65) and parents of children with Down syndrome (20). Next, some psychometric and machine learning methods were implemented to select the items having more power to discriminate children with autism from normal children and children with Down syndrome and evaluate its reliability and validity features. Results: Findings yielded good reliability (0.96) and criterion validity (0.86) for the system. The accuracy was about 0.93 and .92, using Random Forest (RF) and Support Vector Machines (SVM)), respectively. In addition, specificity and sensitivity of the system using SVM is 84.1 and 98.5%, respectively, while RF is 73.4 specificity and 99.3% sensitivity. This system can be considered as a reliable and valid system for screening ASDs.

2015 3rd RSI International Conference on Robotics and Mechatronics (ICROM), 2015

The self-reconfiguration planning has been considered as a major problem in the field of modular ... more The self-reconfiguration planning has been considered as a major problem in the field of modular robots. In this paper, we propose a human-inspired map-based rout planning approach that speeds up the reconfiguration planning. In this method, the analogous to a map is a reconfiguration hub-graph which contains the major nodes of the reconfiguration graph. We propose an RRT-based approach to select these hubs such that they efficiently cover the full configuration graph. In a search for a path between two nodes of the graph, a pre-calculated optimal routing from a hub to another hub may be used. This method has been simulated for two different modular robots and the results show that, the time spent to find a solution is independent of the solution length. Experiments show 5 to 8 fold speed up in search time, while founded solutions are near-optimal. Also it has been shown that the configuration graph is scale free therefore the idea of using hubs is extendable to larger number of modules.