- Robot Manipulation and Learning
- Social Robot Interaction and HRI
- Robotic Path Planning Algorithms
- Teleoperation and Haptic Systems
- Robotics and Sensor-Based Localization
- Augmented Reality Applications
- Human-Automation Interaction and Safety
- AI in Service Interactions
- Hand Gesture Recognition Systems
- Evacuation and Crowd Dynamics
- Robotics and Automated Systems
- Interactive and Immersive Displays
- Virtual Reality Applications and Impacts
- Human Pose and Action Recognition
- Robotic Locomotion and Control
- Action Observation and Synchronization
- Motor Control and Adaptation
- Gaze Tracking and Assistive Technology
- Tactile and Sensory Interactions
- Prosthetics and Rehabilitation Robotics
- Human Mobility and Location-Based Analysis
- Face Recognition and Perception
- Mobile Crowdsensing and Crowdsourcing
- Reinforcement Learning in Robotics
- Manufacturing Process and Optimization
Monash University
2020-2025
Australian Regenerative Medicine Institute
2020-2023
Institute of Electrical and Electronics Engineers
2022
Gorgias Press (United States)
2022
Stevens Institute of Technology
2022
University of British Columbia
2012-2022
Engineering Systems (United States)
2020
The University of Tokyo
2013-2017
Tokyo University of Information Sciences
2013-2015
Washington University in St. Louis
2011
This paper presents a future-focused approach for robot programming based on augmented trajectories. Using mixed reality head-mounted display (Microsoft Hololens) and 7-DOF arm, we designed an (AR) robotic interface with four interactive functions to ease the task: 1) Trajectory specification. 2) Virtual previews of motion. 3) Visualization parameters. 4) Online reprogramming during simulation execution. We validate our AR-robot teaching by comparing it kinesthetic in two different scenarios...
In this study, we investigate and characterize haptic interaction in human-to-human handovers identify key features that facilitate safe efficient object transfer. Eighteen participants worked pairs transferred weighted objects to each other while measured their grip forces load forces. Our data show during transfer, both the giver receiver employ a similar strategy for controlling response changes addition, an implicit social contract appears exist which is responsible ensuring safety...
In this paper, we present a novel controller for safe, efficient, and intuitive robot-to-human object handovers set of experiments to evaluate user responses the robot’s handover behavior. The enables robot mimic human behavior by actively regulating applied grip force according measured load during handover. We provide an implementation on Willow Garage PR2 robot, demonstrating feasibility realizing our design robots with basic sensor/actuator capabilities. A study comparing four variations...
We provide an experimental evaluation of a wearable augmented reality (AR) system we have developed for human-robot teams working on tasks requiring collaboration in shared physical workspace. Recent advances AR technology facilitated the development more intuitive user interfaces many interaction applications. While it has been anticipated that can interface to robot assistants helping human workers various manufacturing scenarios, existing studies robotics largely limited teleoperation and...
Humans are highly skilled in communicating their intent for when and where a handover would occur. However, even the state-of-the-art robotic implementations handovers typically lack of such communication skills. This study investigates visualization robot's internal state Human-to-Robot Handovers using Augmented Reality. Specifically, we explore use visualized 3D models object gripper to communicate estimation is pose which robot intends grasp object. We tested this design via user with 16...
When humans work together to complete a joint task, each person builds an internal model of the situation and how it will evolve. Efficient collaboration depends on these individual models overlap form shared mental among team members; are also important for collaborative processes in human-robot teams. The development maintenance accurate requires bidirectional communication intent ability interpret other members. To enable effective collaboration, this paper investigates use augmented...
Safety is a basic requirement in any human-robot collaboration scenario. To ensure user safety, from both physical and psychological aspects, we propose novel Virtual Barrier system facilitated by an augmented reality interface <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> . Our provides two types of Barriers to safety: 1) Person which encapsulates follows the protect them collisions with robot, 2) Obstacle users can spawn objects or...
To enable robots to learn handover orientations from observing natural handovers, we conduct a user study measure and compare with giver-centered receiver-centered for twenty common objects. We use distance minimization approach compute mean orientations. posit that, computed means of could be used by robot givers achieve more efficient socially acceptable handovers. Furthermore, introduce the notion affordance axes comparing orientations, offer definition computing them. Observable patterns...
In this paper, we propose a novel navigation system for mobile robots in pedestrian-rich sidewalk environments. Sidewalks are unique that the pedestrian-shared space has characteristics of both roads and indoor spaces. Like vehicles on roads, pedestrian movement often manifests as linear flows opposing directions. On other hand, pedestrians also form crowds can exhibit much more random movements than vehicles. Classical algorithms insufficient safe around remaining space. Thus, our approach...
This paper presents the design, implementation and experimental evaluation of a semi-humanoid robotic system for autonomous multi-floor navigation. robot, Personal Robot 2 named Charlie, is capable operating an elevator to travel between rooms located on separate floors. Our goal create assistant locating points interest, manipulating objects, navigating in multi-storied environment equipped with elevator. Taking requires robot (1) map localize within its environment, (2) navigate door, (3)...
This paper investigates the use of kinematic motions recognized by a support vector machine (SVM) for automatic detection object handovers from perspective an receiver. The classifier uses giver’s behaviors (e.g. joint angles, distances joints each other and with respect to receiver) determine intent hand over object. We used bagged random forest how informative features were in predicting occurrence handovers, assist selecting core set be classifier. Altogether, 22 chosen developing...
We present a method for enabling robots to determine appropriate grasp configurations handovers - i.e., where grasp, and how orient an object when handing it over. In our method, robot first builds knowledge base by observing demonstrations of certain objects are used their proper handover configurations. Objects in the then organized based on movements inter-object interaction features. The key point this process is that similarity affordances should be recognized. When subsequently asked...
This paper investigate potential use of augmented reality (AR) for physical human-robot collaboration in large-scale, labour-intensive manufacturing tasks. While it has been shown that AR can help increase task efficiency teleoperative and robot programming tasks involving smaller-scale robots, its shared workspaces large-scale have not well-studied. With the eventual goal applying our system to collaborative aircraft body manufacturing, we compare a user study an interface developed with...
Current robot interfaces, such as teach pendants and 2D screen displays, used for task visualization interaction often seem unintuitive limited in terms of information flow. This compromises efficiency interacting with the interface can distract user from at hand. Augmented reality (AR) technology offers capability to create visually rich displays intuitive elements situ. In recent years, AR has shown promising potential enable effective human–robot (HRI). We introduce ARviz, a versatile,...
We present an approach for visualizing mobile robots through Augmented Reality headset when there is no line-of-sight visibility between the robot and human. Three elements are visualized in Reality: 1) Robot's 3D model to indicate its position, 2) An arrow emanating from planned movement direction, 3) A 2D grid represent ground plane. conduct a user study with 18 participants, which each participant asked retrieve objects, one at time, stations two sides of T-junction end hallway where...
Recent protocols and metrics for training evaluating autonomous robot navigation through crowds are inconsistent due to diversified definitions of "social behavior". This makes it difficult, if not impossible, effectively compare published algorithms. Without a good evaluation protocol resulting algorithms may fail generalize, lack diversity in training. To address these gaps, this paper facilitates more comprehensive objective comparison crowd by proposing consistent set that accounts both...
The increasing presence of robots alongside humans, such as in human-robot teams manufacturing, gives rise to research questions about the kind behaviors people prefer their robot counterparts. We term actions that support interaction by reducing future interference with others supportive and investigate utility a co-located manipulation scenario. compare two modes shared table pick-and-place task: (1) Task-oriented: only takes further its task objective (2) Supportive: sometimes prefers...
As humanoids work alongside people, there will be many situations where they need to handover objects people. If are fulfill their purpose effectively, it is imperative that perform handovers properly. When handing over an object, the giver needs determine grasp and how orient object properly in order ensure a safe efficient handover. We propose implement framework for automatically learning points orientations - which we refer collectively as configuration by observing people hand robot....
An appropriate user interface to collect human demonstration data for deformable object manipulation has been mostly overlooked in the literature. We present an inter-action design demonstrating cloth folding robots. Users choose pick and place points on can preview a visualization of simulated before real-robot execution. Two interfaces are proposed: A 2D display-and-mouse where placed by clicking image cloth, 3D Augmented Reality chosen hand gestures. conduct study with 18 participants,...
Existing approaches to direct robot-to-human handovers are typically implemented on fixed-base robot arms, or mobile manipulators that come a full stop before performing the handover. We propose "on-the-go" which permit moving manipulator hand over an object human without stopping. The on-the-go handover motion is generated with reactive controller allows simultaneous control of base and arm. In user study, receivers subjectively assessed be more efficient, predictable, natural, better timed...
Object handover is a basic task in many human-robot interactive scenarios and therefore, it important for assistive robots to be able perform proper handovers. We previously designed human-inspired grip-force-varying controller robot giver showed on Willow Garage PR2 that the yields human-like human-preferred The had non-compliant fully-actuated gripper. However, recently, compliant underactuated grippers have been gaining more popularity. Although can provide flexibility manipulation,...
As robots have become increasingly common in human-rich environments, it is critical that they are able to exhibit social cues be perceived as a cooperative and socially-conformant team member. We investigate the effect of robot gaze on people's subjective perceptions mobile socially present entity three hallway navigation scenarios. The tested behaviors were path-oriented (looking at its own future path), or person-oriented nearest person), with fixed-gaze control. conduct real-world study...