Robot navigation in the presence of humans raises new issues for motion planning and control sinc... more Robot navigation in the presence of humans raises new issues for motion planning and control since the humans safety and comfort must be taken explicitly into account. We claim that a human-aware motion planner must not only elaborate safe robot paths, but also plan good, socially acceptable and legible paths. Our aim is to build a planner that takes explicitly into account the human partner by reasoning about his accessibility, his vision field and potential shared motions. This paper focuses on a navigation planner that takes into account the humans existence explicitly. This planner is part of a human-aware motion and manipulation planning and control system that we aim to develop in order to achieve motion and manipulation tasks in a collaborative way with the human. We are conducting research in a multidisciplinary perspective, (1) running user studies and (2) developing an algorithmic framework able to integrate knowledge acquired through the trials. We illustrate here a first step by implementing a humanfriendly approach motion by the robot.
Identifying the roles and the specific social behaviours that evoke human trust towards robots is... more Identifying the roles and the specific social behaviours that evoke human trust towards robots is key for user acceptance. Specially, while performing tasks in the real world, such as navigation or guidance, the predictability of robot motion and predictions of user intentions facilitate interaction. We present a user study in which a humanoid-robot guided participants around a human populated environment, avoiding collisions while following a socially acceptable trajectory. We investigated which behaviours performed by a humanoid robot during a guidance task exhibited better social acceptance by people, and how robot behaviours influence their trust in a robot to safely complete a guiding task. We concluded that in general, people prefer and trust a robot that exhibits social behaviours such as talking and maintaining an appropriate safe distance from obstacles.
Outcomes are presented from experiments on the effect of participants' individual preferences for... more Outcomes are presented from experiments on the effect of participants' individual preferences for robot appearance and height on their preferences towards and perceptions of live robots. Participants who expressed a preference for a mechanical looking robot, tended to prefer all robot types to stay further away than those participants who expressed a preference for more humanoid robots. A majority group of two thirds (68.5%) preferred a robot which they personally perceived as having an extrovert and agreeable personality and a minority third (31.5%) preferred no strong robot personality factors. Humanoid robots also tended to be perceived as more intelligent than the mechanoid robots, but when combined with short height, were seen as less conscientious and more neurotic. The taller robots overall were also perceived as more human-like and conscientious than the short robots.
The propose of this study was to assess the feasibility of using support vector machines in analy... more The propose of this study was to assess the feasibility of using support vector machines in analysing myoelectric signals acquired using an off the shelf device, the Myo armband from Thalmic Lab, when performing hand grasp gestures. Participants (n = 26) took part in the study wearing the armband and producing a series of required gestures. Support vector machines were used to train a model using participant training values, and to classify gestures produced by the same participants. Different Kernel functions and electrode combinations were studied. Also we contrasted different lengths of training values versus different lengths for the classification samples. The overall accuracy was 94.9% with data from 8 electrodes, and 72% where only four of the electrodes were used. The linear kernel outperformed the polynomial, and radial basis function. Exploring the number of training samples versus the achieved classification accuracy, results identified acceptable accuracies (> 90%) for training around 2.5s, and recognising grasp with 0.2s of acquired data. The best recognised grasp was the hand closed (97.6%), followed by cylindrical grasp (96.8%), the lateral grasp (93.2%) and tripod (92%). These results allows us to progress to the next stage of work where the Myo armband is used in the context of robot-mediated stroke rehabilitation and also involves more dynamic interactions as well as gross upper arm movements.
Trust is a key factor in human users' acceptance of robots in a home or human oriented environmen... more Trust is a key factor in human users' acceptance of robots in a home or human oriented environment. Humans should be able to trust that they can safely interact with their robot. Robots will sometimes make errors, due to mechanical or functional failures. It is therefore important that a domestic robot should have acceptable interactive behaviours when exhibiting and recovering from an error situation. In order to define these behaviours, it is firstly necessary to consider that errors can have different degrees of consequences. We hypothesise that the severity of the consequences and the timing of a robot's different types of erroneous behaviours during an interaction may have different impacts on users' attitudes towards a domestic robot. In this study we used an interactive storyboard presenting ten different scenarios in which a robot performed different tasks under five different conditions. Each condition included the ten different tasks performed by the robot, either correctly, or with small or big errors. The conditions with errors were complemented with four correct behaviours. At the end of each experimental condition, participants were presented with an emergency scenario to evaluate their current trust in the robot. We conclude that there is correlation between the magnitude of an error performed by the robot and the corresponding loss of trust of the human in the robot.
The propose of this study was to assess the feasibility of using support vector machines in analy... more The propose of this study was to assess the feasibility of using support vector machines in analysing myoelectric signals acquired using an off the shelf device, the Myo armband from Thalmic Lab.
Human-robot interaction is a rapidly growing research area which more and more roboticists and co... more Human-robot interaction is a rapidly growing research area which more and more roboticists and computer scientists are moving into. Publications on work resulting from such studies rarely consider in detail the practical and methodological problems encountered. This paper aims to highlight and critically discuss such problems involved in conducting human-robot interaction studies. We provide some examples by discussing our experiences of running two trials that involved humans and robots physically interacting in a common space. Our discussion emphasises the need to take safety requirements into account, and minimise the risk of physical harm to human subjects. Ethical considerations are considered, which are often within a formal or legal framework depending on the host country or institution. We also discuss future improvements for features of our trials and make suggestions as to how to overcome the challenges we encountered. We hope that the lessons learnt will be used to improve future human-robot interaction trials.
The main aim of this study was to confirm the have also conducted similar EiRT trials with human ... more The main aim of this study was to confirm the have also conducted similar EiRT trials with human sized findings from previous pilot studies that results obtained from robots including Dario et al. [7], Severinson-Eklundh et al. the same Human Robot Interaction (HRI) scenarios in trials [8], Kanda et al. [9] and Hinds et al. [10]. However, most of using both video-based and live methodologies were these HiRI trials to date typically are characterised by comparable. We investigated how a robot should approach relatively small sample sizes and are highly exploratory. HRI human subjects in various scenarios relevant to the robot
The results from two empirical studies of human-robot interaction are presented. The first study ... more The results from two empirical studies of human-robot interaction are presented. The first study involved the subject approaching the static robot and the robot approaching the standing subject. In these trials a small majority of subjects preferred a distance corresponding to the 'personal zone' typically used by humans when talking to friends. However, a large minority of subjects got significantly closer, suggesting that they treated the robot differently from a person, and possibly did not view the robot as a social being. The second study involved a scenario where the robot fetched an object that the seated subject had requested, arriving from different approach directions. The results of this second trial indicated that most subjects disliked a frontal approach. Most subjects preferred to be approached from either the left or right side, with a small overall preference for a right approach by the robot. Implications for future work are discussed.
Identifying links between human personality and attributed robot personality is a relatively new ... more Identifying links between human personality and attributed robot personality is a relatively new area of human–robot interaction. In this paper we report on an exploratory study that investigates human and robot personality traits as part of a human–robot interaction trial. The trials took place in a simulated living-room scenario involving 28 participants and a human-sized robot of mechanical appearance. Participants interacted with the robot in two task scenarios relevant to a ‘robot in the home’ context. It was found that participants’ evaluations of their own personality traits are related to their evaluations of the robot’s personality traits. The statistical analysis of questionnaire data yields several statistically significant results: (a) Participants do not tend to assign their personality traits to match the robots’, (b) For individual personality traits, participants rated themselves as having stronger personality characteristics compared to the robot, (c) Specific significant correlations were found between participants’ and robot personality traits, and (d) Significant group differences for participant gender, age and technological background are highlighted. The results are discussed in light of developing personalized robot companions.
In this work we investigate how humans' awareness of a social robot's capabilities affect their t... more In this work we investigate how humans' awareness of a social robot's capabilities affect their trust in the robot. We present a user study that relates knowledge on different quality levels to participants' ratings of trust. Primary school pupils were asked to rate their trust in the robot after three types of interactions: a video demonstration, a live interaction, and a programming task. The study revealed that the pupils' trust is not significantly affected across different domains after each session. It did not appear to be significant differences in trust tendencies for the different experiences either; however, our results suggest that human users trust a robot more the more awareness about the robot they have.
As we expect that the presence of autonomous robots in our everyday life will increase, we must c... more As we expect that the presence of autonomous robots in our everyday life will increase, we must consider that people will have not only to accept robots to be a fundamental part of their lives, but they will also have to trust them to reliably and securely engage them in collaborative tasks. Several studies showed that robots are more comfortable interacting with robots that respect social conventions. However, it is still not clear if a robot that expresses social conventions will gain more favourably people’s trust. In this study, we aimed to assess whether the use of social behaviours and natural communications can affect humans’ sense of trust and companionship towards the robots. We conducted a between-subjects study where participants’ trust was tested in three scenarios with increasing trust criticality (low, medium, high) in which they interacted either with a social or a non-social robot. Our findings showed that participants trusted equally a social and non-social robot in the low and medium consequences scenario. On the contrary, we observed that participants’ choices of trusting the robot in a higher sensitive task was affected more by a robot that expressed social cues with a consequent decrease of their trust in the robot.
Trust has been established to be a key factor in fostering human-robot interactions. However, tru... more Trust has been established to be a key factor in fostering human-robot interactions. However, trust can change overtime according to different factors, including a breach of trust due to a robot's error. In this exploratory study, we observed people's interactions with a companion robot in a real house, adapted for human-robot interaction experimentation, over three weeks. The interactions happened in six scenarios in which a robot performed different tasks under two different conditions. Each condition included fourteen tasks performed by the robot, either correctly, or with errors with severe consequences on the first or last day of interaction. At the end of each experimental condition, participants were presented with an emergency scenario to evaluate their trust in the robot. We evaluated participants' trust in the robot by observing their decision to trust the robot during the emergency scenario, and by collecting their views through questionnaires. We concluded that there is a correlation between the timing of an error with severe consequences performed by the robot and the corresponding loss of trust of the human in the robot. In particular, people's trust is subjected to the initial mental formation.
This paper describes an extensive analysis of the comfort level data of 7 subjects with respect t... more This paper describes an extensive analysis of the comfort level data of 7 subjects with respect to 12 robot behaviours as part of a human-robot interaction trial. This includes robot action, proximity and motion relative to the subjects. Two researchers coded the video material, identifying visible states of discomfort displayed by subjects in relation to the robot's behaviour. Agreement between the coders varied from moderate to high, except for more ambiguous situations involving robot approach directions. The detected visible states of discomfort were correlated with the situations where the comfort level device (CLD) indicated states of discomfort. Results show that the uncomfortable states identified by both coders, and by either of the coders corresponded with 31% and 64% of the uncomfortable states identified by the subjects' CLD data (N=58), respectively. Conversely there was 72% agreement between subjects' CLD data and the uncomfortable states identified by both coders (N=25). Results show that the majority of the subjects expressed discomfort when the robot blocked their path or was on a collision course towards them, especially when the robot was within 3 meters proximity. Other observations include that the majority of subjects experienced discomfort when the robot was closer than 3m, within the social zone reserved for human-human face to face conversation, while they were performing a task. The advantages and disadvantages of the CLD in comparison to other techniques for assessing subjects' internal states are discussed and future work concludes the paper.
This paper presents results of a pilot study that investigated whether people's perceptions from ... more This paper presents results of a pilot study that investigated whether people's perceptions from live and video HRI trials were comparable. Subjects participated in a live HRI trial and videotaped HRI trials in which the scenario for both trials was identical, and involved a robot fetching an object using different approach directions. Results of the trials indicated moderate to high levels of agreement for subjects' preferences, and opinions for both the live and video based HRI trials. This methodology is in its infancy and should not be seen as a replacement for live trials. However, our results indicate that for certain HRI scenarios videotaped trials do have potential as a technique for prototyping, testing, developing HRI scenarios, and testing methodologies for use in definitive live trials.
The propose of this study was to assess the feasibility of using myoelectric signals acquired usi... more The propose of this study was to assess the feasibility of using myoelectric signals acquired using an off the shelf device, the Myo armband from Thalmic Lab. Background: With the technological advances in sensing human motion, and its potential to drive and control mechanical interfaces remotely, a multitude of input mechanisms are used to link actions between the human and the robot. In this study we explored the feasibility of using human arm's myoelectric signals with the aim of identifying a number of gestures automatically. Material and methods: Participants (n = 26) took part in a study with the aim to assess the gesture detection accuracy using myoelectric signals. The Myo armband was used worn on the forearm. The session was divided into three phases, familiarisation: where participant learned how to use the armband, training: when participants reproduced a number of requested gestures to train our machine learning algorithm and recognition: when gestures presented on screen where reproduced by participants, and simultaneously recognised using the machine learning routines. Results: One participant did not complete the study due to technical errors during the session. The remaining (n = 25) participants completed the study allowing to calculate individual accuracy for grasp detection using this medium. Our overall accuracy was 65.06%, with the cylindrical grasp achieving the
As robots increasingly take part in daily living activities, humans will have to interact with th... more As robots increasingly take part in daily living activities, humans will have to interact with them in domestic and other human-oriented environments. We can expect that domestic robots will exhibit occasional mechanical, programming or functional errors, as occur with other electrical consumer devices. For example, these errors could include software errors, dropping objects due to gripper malfunctions, picking up the wrong object or showing faulty navigational skills due to unclear camera images or noisy laser scanner data respectively. It is therefore important for a domestic robot to have acceptable interactive behaviour when exhibiting and recovering from an error situation. As a first step, the current study investigated human users’ perceptions of the severity of various categories of potential errors that are likely to be exhibited by a domestic robot. We conducted a questionnaire-based study, where participants rated 20 different scenarios in which a domestic robot made an error. The potential errors were rated by participants by severity. Our findings indicate that people perceptions of the magnitude of the errors presented in the questionnaire were consistent. We did not find any significant differences in users’ ratings due to age and gender. We clearly identified scenarios that were rated by participants as having limited consequences (“small” errors) and that were rated as having severe consequences (“big” errors). Future work will use these two sets of consistently rated robot error scenarios as baseline scenarios to perform studies with repeated interactions investigating human perceptions of robot tasks and error severity.
This paper presents an Autonomous Proxemic System (APS) for a mobile robot. It detects people in ... more This paper presents an Autonomous Proxemic System (APS) for a mobile robot. It detects people in the surroundings and manipulates the robot's motions to approach them keeping an acceptable proxemic distance. The APS sensing functions include face and upper body detection, leg detection, and motion detection using camera, laser, and infra-red sensors respectively. The control functions consist of approach a human and obstacle avoidance. APS uses the sonar and laser range devices to keep an accurate proxemic distance with the human. Initial system tests indicate that the APS keeps desired proxemic distances to within an acceptable error margin.
Robot navigation in the presence of humans raises new issues for motion planning and control sinc... more Robot navigation in the presence of humans raises new issues for motion planning and control since the humans safety and comfort must be taken explicitly into account. We claim that a human-aware motion planner must not only elaborate safe robot paths, but also plan good, socially acceptable and legible paths. Our aim is to build a planner that takes explicitly into account the human partner by reasoning about his accessibility, his vision field and potential shared motions. This paper focuses on a navigation planner that takes into account the humans existence explicitly. This planner is part of a human-aware motion and manipulation planning and control system that we aim to develop in order to achieve motion and manipulation tasks in a collaborative way with the human. We are conducting research in a multidisciplinary perspective, (1) running user studies and (2) developing an algorithmic framework able to integrate knowledge acquired through the trials. We illustrate here a first step by implementing a humanfriendly approach motion by the robot.
Identifying the roles and the specific social behaviours that evoke human trust towards robots is... more Identifying the roles and the specific social behaviours that evoke human trust towards robots is key for user acceptance. Specially, while performing tasks in the real world, such as navigation or guidance, the predictability of robot motion and predictions of user intentions facilitate interaction. We present a user study in which a humanoid-robot guided participants around a human populated environment, avoiding collisions while following a socially acceptable trajectory. We investigated which behaviours performed by a humanoid robot during a guidance task exhibited better social acceptance by people, and how robot behaviours influence their trust in a robot to safely complete a guiding task. We concluded that in general, people prefer and trust a robot that exhibits social behaviours such as talking and maintaining an appropriate safe distance from obstacles.
Outcomes are presented from experiments on the effect of participants' individual preferences for... more Outcomes are presented from experiments on the effect of participants' individual preferences for robot appearance and height on their preferences towards and perceptions of live robots. Participants who expressed a preference for a mechanical looking robot, tended to prefer all robot types to stay further away than those participants who expressed a preference for more humanoid robots. A majority group of two thirds (68.5%) preferred a robot which they personally perceived as having an extrovert and agreeable personality and a minority third (31.5%) preferred no strong robot personality factors. Humanoid robots also tended to be perceived as more intelligent than the mechanoid robots, but when combined with short height, were seen as less conscientious and more neurotic. The taller robots overall were also perceived as more human-like and conscientious than the short robots.
The propose of this study was to assess the feasibility of using support vector machines in analy... more The propose of this study was to assess the feasibility of using support vector machines in analysing myoelectric signals acquired using an off the shelf device, the Myo armband from Thalmic Lab, when performing hand grasp gestures. Participants (n = 26) took part in the study wearing the armband and producing a series of required gestures. Support vector machines were used to train a model using participant training values, and to classify gestures produced by the same participants. Different Kernel functions and electrode combinations were studied. Also we contrasted different lengths of training values versus different lengths for the classification samples. The overall accuracy was 94.9% with data from 8 electrodes, and 72% where only four of the electrodes were used. The linear kernel outperformed the polynomial, and radial basis function. Exploring the number of training samples versus the achieved classification accuracy, results identified acceptable accuracies (> 90%) for training around 2.5s, and recognising grasp with 0.2s of acquired data. The best recognised grasp was the hand closed (97.6%), followed by cylindrical grasp (96.8%), the lateral grasp (93.2%) and tripod (92%). These results allows us to progress to the next stage of work where the Myo armband is used in the context of robot-mediated stroke rehabilitation and also involves more dynamic interactions as well as gross upper arm movements.
Trust is a key factor in human users' acceptance of robots in a home or human oriented environmen... more Trust is a key factor in human users' acceptance of robots in a home or human oriented environment. Humans should be able to trust that they can safely interact with their robot. Robots will sometimes make errors, due to mechanical or functional failures. It is therefore important that a domestic robot should have acceptable interactive behaviours when exhibiting and recovering from an error situation. In order to define these behaviours, it is firstly necessary to consider that errors can have different degrees of consequences. We hypothesise that the severity of the consequences and the timing of a robot's different types of erroneous behaviours during an interaction may have different impacts on users' attitudes towards a domestic robot. In this study we used an interactive storyboard presenting ten different scenarios in which a robot performed different tasks under five different conditions. Each condition included the ten different tasks performed by the robot, either correctly, or with small or big errors. The conditions with errors were complemented with four correct behaviours. At the end of each experimental condition, participants were presented with an emergency scenario to evaluate their current trust in the robot. We conclude that there is correlation between the magnitude of an error performed by the robot and the corresponding loss of trust of the human in the robot.
The propose of this study was to assess the feasibility of using support vector machines in analy... more The propose of this study was to assess the feasibility of using support vector machines in analysing myoelectric signals acquired using an off the shelf device, the Myo armband from Thalmic Lab.
Human-robot interaction is a rapidly growing research area which more and more roboticists and co... more Human-robot interaction is a rapidly growing research area which more and more roboticists and computer scientists are moving into. Publications on work resulting from such studies rarely consider in detail the practical and methodological problems encountered. This paper aims to highlight and critically discuss such problems involved in conducting human-robot interaction studies. We provide some examples by discussing our experiences of running two trials that involved humans and robots physically interacting in a common space. Our discussion emphasises the need to take safety requirements into account, and minimise the risk of physical harm to human subjects. Ethical considerations are considered, which are often within a formal or legal framework depending on the host country or institution. We also discuss future improvements for features of our trials and make suggestions as to how to overcome the challenges we encountered. We hope that the lessons learnt will be used to improve future human-robot interaction trials.
The main aim of this study was to confirm the have also conducted similar EiRT trials with human ... more The main aim of this study was to confirm the have also conducted similar EiRT trials with human sized findings from previous pilot studies that results obtained from robots including Dario et al. [7], Severinson-Eklundh et al. the same Human Robot Interaction (HRI) scenarios in trials [8], Kanda et al. [9] and Hinds et al. [10]. However, most of using both video-based and live methodologies were these HiRI trials to date typically are characterised by comparable. We investigated how a robot should approach relatively small sample sizes and are highly exploratory. HRI human subjects in various scenarios relevant to the robot
The results from two empirical studies of human-robot interaction are presented. The first study ... more The results from two empirical studies of human-robot interaction are presented. The first study involved the subject approaching the static robot and the robot approaching the standing subject. In these trials a small majority of subjects preferred a distance corresponding to the 'personal zone' typically used by humans when talking to friends. However, a large minority of subjects got significantly closer, suggesting that they treated the robot differently from a person, and possibly did not view the robot as a social being. The second study involved a scenario where the robot fetched an object that the seated subject had requested, arriving from different approach directions. The results of this second trial indicated that most subjects disliked a frontal approach. Most subjects preferred to be approached from either the left or right side, with a small overall preference for a right approach by the robot. Implications for future work are discussed.
Identifying links between human personality and attributed robot personality is a relatively new ... more Identifying links between human personality and attributed robot personality is a relatively new area of human–robot interaction. In this paper we report on an exploratory study that investigates human and robot personality traits as part of a human–robot interaction trial. The trials took place in a simulated living-room scenario involving 28 participants and a human-sized robot of mechanical appearance. Participants interacted with the robot in two task scenarios relevant to a ‘robot in the home’ context. It was found that participants’ evaluations of their own personality traits are related to their evaluations of the robot’s personality traits. The statistical analysis of questionnaire data yields several statistically significant results: (a) Participants do not tend to assign their personality traits to match the robots’, (b) For individual personality traits, participants rated themselves as having stronger personality characteristics compared to the robot, (c) Specific significant correlations were found between participants’ and robot personality traits, and (d) Significant group differences for participant gender, age and technological background are highlighted. The results are discussed in light of developing personalized robot companions.
In this work we investigate how humans' awareness of a social robot's capabilities affect their t... more In this work we investigate how humans' awareness of a social robot's capabilities affect their trust in the robot. We present a user study that relates knowledge on different quality levels to participants' ratings of trust. Primary school pupils were asked to rate their trust in the robot after three types of interactions: a video demonstration, a live interaction, and a programming task. The study revealed that the pupils' trust is not significantly affected across different domains after each session. It did not appear to be significant differences in trust tendencies for the different experiences either; however, our results suggest that human users trust a robot more the more awareness about the robot they have.
As we expect that the presence of autonomous robots in our everyday life will increase, we must c... more As we expect that the presence of autonomous robots in our everyday life will increase, we must consider that people will have not only to accept robots to be a fundamental part of their lives, but they will also have to trust them to reliably and securely engage them in collaborative tasks. Several studies showed that robots are more comfortable interacting with robots that respect social conventions. However, it is still not clear if a robot that expresses social conventions will gain more favourably people’s trust. In this study, we aimed to assess whether the use of social behaviours and natural communications can affect humans’ sense of trust and companionship towards the robots. We conducted a between-subjects study where participants’ trust was tested in three scenarios with increasing trust criticality (low, medium, high) in which they interacted either with a social or a non-social robot. Our findings showed that participants trusted equally a social and non-social robot in the low and medium consequences scenario. On the contrary, we observed that participants’ choices of trusting the robot in a higher sensitive task was affected more by a robot that expressed social cues with a consequent decrease of their trust in the robot.
Trust has been established to be a key factor in fostering human-robot interactions. However, tru... more Trust has been established to be a key factor in fostering human-robot interactions. However, trust can change overtime according to different factors, including a breach of trust due to a robot's error. In this exploratory study, we observed people's interactions with a companion robot in a real house, adapted for human-robot interaction experimentation, over three weeks. The interactions happened in six scenarios in which a robot performed different tasks under two different conditions. Each condition included fourteen tasks performed by the robot, either correctly, or with errors with severe consequences on the first or last day of interaction. At the end of each experimental condition, participants were presented with an emergency scenario to evaluate their trust in the robot. We evaluated participants' trust in the robot by observing their decision to trust the robot during the emergency scenario, and by collecting their views through questionnaires. We concluded that there is a correlation between the timing of an error with severe consequences performed by the robot and the corresponding loss of trust of the human in the robot. In particular, people's trust is subjected to the initial mental formation.
This paper describes an extensive analysis of the comfort level data of 7 subjects with respect t... more This paper describes an extensive analysis of the comfort level data of 7 subjects with respect to 12 robot behaviours as part of a human-robot interaction trial. This includes robot action, proximity and motion relative to the subjects. Two researchers coded the video material, identifying visible states of discomfort displayed by subjects in relation to the robot's behaviour. Agreement between the coders varied from moderate to high, except for more ambiguous situations involving robot approach directions. The detected visible states of discomfort were correlated with the situations where the comfort level device (CLD) indicated states of discomfort. Results show that the uncomfortable states identified by both coders, and by either of the coders corresponded with 31% and 64% of the uncomfortable states identified by the subjects' CLD data (N=58), respectively. Conversely there was 72% agreement between subjects' CLD data and the uncomfortable states identified by both coders (N=25). Results show that the majority of the subjects expressed discomfort when the robot blocked their path or was on a collision course towards them, especially when the robot was within 3 meters proximity. Other observations include that the majority of subjects experienced discomfort when the robot was closer than 3m, within the social zone reserved for human-human face to face conversation, while they were performing a task. The advantages and disadvantages of the CLD in comparison to other techniques for assessing subjects' internal states are discussed and future work concludes the paper.
This paper presents results of a pilot study that investigated whether people's perceptions from ... more This paper presents results of a pilot study that investigated whether people's perceptions from live and video HRI trials were comparable. Subjects participated in a live HRI trial and videotaped HRI trials in which the scenario for both trials was identical, and involved a robot fetching an object using different approach directions. Results of the trials indicated moderate to high levels of agreement for subjects' preferences, and opinions for both the live and video based HRI trials. This methodology is in its infancy and should not be seen as a replacement for live trials. However, our results indicate that for certain HRI scenarios videotaped trials do have potential as a technique for prototyping, testing, developing HRI scenarios, and testing methodologies for use in definitive live trials.
The propose of this study was to assess the feasibility of using myoelectric signals acquired usi... more The propose of this study was to assess the feasibility of using myoelectric signals acquired using an off the shelf device, the Myo armband from Thalmic Lab. Background: With the technological advances in sensing human motion, and its potential to drive and control mechanical interfaces remotely, a multitude of input mechanisms are used to link actions between the human and the robot. In this study we explored the feasibility of using human arm's myoelectric signals with the aim of identifying a number of gestures automatically. Material and methods: Participants (n = 26) took part in a study with the aim to assess the gesture detection accuracy using myoelectric signals. The Myo armband was used worn on the forearm. The session was divided into three phases, familiarisation: where participant learned how to use the armband, training: when participants reproduced a number of requested gestures to train our machine learning algorithm and recognition: when gestures presented on screen where reproduced by participants, and simultaneously recognised using the machine learning routines. Results: One participant did not complete the study due to technical errors during the session. The remaining (n = 25) participants completed the study allowing to calculate individual accuracy for grasp detection using this medium. Our overall accuracy was 65.06%, with the cylindrical grasp achieving the
As robots increasingly take part in daily living activities, humans will have to interact with th... more As robots increasingly take part in daily living activities, humans will have to interact with them in domestic and other human-oriented environments. We can expect that domestic robots will exhibit occasional mechanical, programming or functional errors, as occur with other electrical consumer devices. For example, these errors could include software errors, dropping objects due to gripper malfunctions, picking up the wrong object or showing faulty navigational skills due to unclear camera images or noisy laser scanner data respectively. It is therefore important for a domestic robot to have acceptable interactive behaviour when exhibiting and recovering from an error situation. As a first step, the current study investigated human users’ perceptions of the severity of various categories of potential errors that are likely to be exhibited by a domestic robot. We conducted a questionnaire-based study, where participants rated 20 different scenarios in which a domestic robot made an error. The potential errors were rated by participants by severity. Our findings indicate that people perceptions of the magnitude of the errors presented in the questionnaire were consistent. We did not find any significant differences in users’ ratings due to age and gender. We clearly identified scenarios that were rated by participants as having limited consequences (“small” errors) and that were rated as having severe consequences (“big” errors). Future work will use these two sets of consistently rated robot error scenarios as baseline scenarios to perform studies with repeated interactions investigating human perceptions of robot tasks and error severity.
This paper presents an Autonomous Proxemic System (APS) for a mobile robot. It detects people in ... more This paper presents an Autonomous Proxemic System (APS) for a mobile robot. It detects people in the surroundings and manipulates the robot's motions to approach them keeping an acceptable proxemic distance. The APS sensing functions include face and upper body detection, leg detection, and motion detection using camera, laser, and infra-red sensors respectively. The control functions consist of approach a human and obstacle avoidance. APS uses the sonar and laser range devices to keep an accurate proxemic distance with the human. Initial system tests indicate that the APS keeps desired proxemic distances to within an acceptable error margin.
Uploads
Papers by Michael Walters