Search

We present a novel method for a robot to interactively learn, while executing, a joint human–robot task. We consider collaborative tasks realized by a team of a human operator and a robot helper that adapts to the human’s task execution preferences. Different human operators can have different abilities, experiences, and personal preferences so that a particular allocation of activities in the team is preferred over another. Our main goal is to have the robot learn the task and the preferences of the user to provide a more efficient and acceptable joint task execution. We cast concurrent multi-agent collaboration as a semi-Markov decision process and show how to model the team behavior and learn the expected robot behavior. We further propose an interactive learning framework and we evaluate it both in simulation and on a real robotic setup to show the system can effectively learn and adapt to human expectations.

Learning human–robot interaction logic from example interaction data has the potential to leverage “big data” to reduce the effort and time spent on designing interaction logic or crafting interaction content. Previous work has demonstrated techniques by which a robot can learn motion and speech behaviors from non-annotated human–human interaction data, but these techniques only enable a robot to respond to human-initiated inputs, and do not enable the robot to proactively initiate interaction. In this work, we propose a method for learning both human-initiated and robot-initiated behavior for a social robot from human–human example interactions, which we demonstrate for a shopkeeper interacting with a customer in a camera shop scenario. This was achieved by extending an existing technique by (1) introducing a concept of a customer yield action, (2) incorporating interaction history, represented by sequences of discretized actions, as inputs for training and generating robot behavior, and (3) using an “attention mechanism” in our learning system for training robot behaviors, that learns which parts of the interaction history are more important for generating robot behaviors. The proposed method trains a robot to generate multimodal actions, consisting of speech and locomotion behaviors. We compared this study with the previous technique in two ways. Cross-validation on the training data showed higher social appropriateness of predicted behaviors using the proposed technique, and a user study of live interaction with a robot showed that participants perceived the proposed technique to produce behaviors that were more proactive, socially-appropriate, and better in overall quality.

The pattern recognition capabilities of deep learning have pushed the limits in various fields—and industrial robotics is no exception. Deep learning alone will not solve all the problems encountered in industrial robotics, but it will certainly improve the perception capabilities of robotics systems, given its power to recognize complex real-world patterns robustly. Here, the author examines robotics applications in deep learning.

Recent advances in machine learning, such as deep neural networks, have caused a huge boost in many different areas of artificial intelligence and robotics. These methods typically require a large corpus of well-prepared and labelled training data, which limits the applicability to robotics. In our opinion, a fundamental challenge in autonomous robotics is to design systems that are simple enough to solve simple tasks. These systems should grow in complexity step by step and more complex models like neural networks should be trained by re-using the information acquired over the robot’s lifetime. Ultimately, high-level abstractions should be generated from these models, bridging the gap from low-level sensor data to high-level AI systems. We present first steps into this direction and analyse their limitations and future extensions in order to achieve the goal of designing autonomous agents.

Numerical P systems (for short, NP systems) are distributed and parallel computing models inspired both from the structure of living cells and from the economic reality, where the values of variables evolve by programs that are composed by production functions and repartition protocols: the value of a production function is distributed to variables according to the corresponding repartition protocol. In this work, we introduce a new method of using evolution programs into NP systems, where thresholds are associated with production functions. The computation power of NP systems with production thresholds is investigated. Specifically, we prove that NP systems with lower-thresholds (the production function value can be distributed only when it is not smaller than a given constant), with one membrane working both in the all-parallel mode and in the sequential mode, are universal. The universal results of NP systems with lower-thresholds are extended to NP systems with upper-thresholds (the production function value can be distributed only when it is not greater than a given constant) by simulating the former with the latter. These universality results show that NP systems with production thresholds have the potential to implement any computer program or robot behavior.

Multimodal communication between humans and autonomous robots is essential to enhance effectiveness of human–robot team performance in complex, novel environments, such as in military intelligence, surveillance, and reconnaissance operations in urban settings. It is imperative that a systematic approach be taken to evaluate the factors that each modality contributes to the user’s ability to perform successfully and safely. This paper addresses the effects of unidirectional speech and gesture methods of communication on perceived workload, usability preferences, and expectations of robot behavior while commanding a robot teammate to perform a spatial-navigation task. Each type of communication was performed alone or simultaneously. Results reveal that although the speech-alone condition elicited the lowest level of perceived workload, the usability preference and expectations of robot behavior after interacting through each communication condition was the same. Further, workload ratings between the gesture and speech-gesture conditions were similar indicating systems that employ gesture communication could also support speech communication with little to no additional subjectively perceived cognitive burden on the user. Findings also reveal that workload alone should not be used as a sole determining factor of communication preference during system and task evaluation and design. Additionally, perceived workload did not seem to negatively impact the level of expectations regarding the robot’s behavior. Recommendations for future human–robot communication evaluation are provided.

This paper presents a robotic behavior planning method under uncertainty based on biology-inspired episodic memory. Adaptive behavior planning, prediction and reasoning are achieved between tasks, environment, and threats. Through building a novel episode model and introducing the activation and stimulation mechanism of state neurons, the framework of an episodic memory-driving Markov decision process (EM-MDP) is proposed for incremental self-learning of robotic experience and cognitive behavior planning. Two main challenges in robot behavior control under uncertainty are addressed: high computational complexity and perceptual aliasing. The approach for robotic global planning and behaviors sequence prediction based on the EM-MDP is developed utilizing neuron synaptic potential. A local behavioral planning method based on risk function and feasible paths is employed to achieve path optimization and behavior reasoning under the condition of imperfect memory. Robot can evaluate the past events sequence, predict the current state, and plan the desired behaviors. The proposed method is evaluated in several real-life environments for a mobile robot system. The robot system is able to successfully produce solutions in general scenarios under uncertainty.

Typically, development of robot behavior entails writing the code, deploying it on a simulator or robot and running it in a test setting. If this feedback reveals errors, the programmer mentally needs to map the error in behavior back to the source code that caused it before being able to fix it. This process suffers from a long cognitive distance between the code and the resulting behavior, which slows down development and can make experimentation with different behaviors prohibitively expensive. In contrast, Live Programming tightens the feedback loop, minimizing the cognitive distance. As a result, programmers benefit from an immediate connection with the program that they are making thanks to an immediate, ‘live’ feedback on program behavior. This allows for extremely rapid creation, or variation, of robot behavior and for dramatically increased debugging speed. To enable such Live Robot Programming, in this article we discuss LRP; our language that provides for live programming of nested state machines. We detail the design of the language and show its features, give an overview of the interpreter and how it enables the liveness properties of the language, and illustrate its independence from specific robot APIs.

Soft, tip-extending devices, or “vine robots,” are a promising new paradigm for navigating cluttered and confined environments. Because they lengthen from their tips, there is little relative movement of the body with the environment, and the compressible nature of the device allows it to pass through orifices smaller than its diameter. However, the interaction between these devices and the environment is not well characterized. Here we present a comprehensive mathematical model that describes vine robot behavior during environmental interaction that provides a basis from which informed designs can be generated in future works. The model incorporates transverse and axial buckling modes that result from growing into obstacles with varying surface normals, as well as internal path-dependent and independent resistances to growth. Accordingly, the model is able to predict the pressure required to grow through a given environment due to the interaction forces it experiences. We experimentally validate both the individual components and the full model. Finally, we present three design insights from the model and demonstrate how they each improve performance in confined space navigation. Our work helps advance the understanding of tip-extending, vine robots through quantifying their interactions with the environment, opening the door for new designs and impactful applications in the realms of healthcare, research, search and rescue, and space exploration.

Humans often assume that robots are rational. We believe robots take optimal actions given their objective; hence, when we are uncertain about what the robot's objective is, we interpret the robot's actions as optimal with respect to our estimate of its objective. This approach makes sense when robots straightforwardly optimize their objective, and enables humans to learn what the robot is trying to achieve. However, our insight is that---when robots are aware that humans learn by trusting that the robot actions are rational---intelligent robots do not act as the human expects; instead, they take advantage of the human's trust, and exploit this trust to more efficiently optimize their own objective. In this paper, we formally model instances of human-robot interaction (HRI) where the human does not know the robot's objective using a two-player game. We formulate different ways in which the robot can model the uncertain human, and compare solutions of this game when the robot has conservative, optimistic, rational, and trusting human models. In an offline linear-quadratic case study and a real-time user study, we show that trusting human models can naturally lead to communicative robot behavior, which influences end-users and increases their involvement.

Social robots progressively enter children’s lives, but little is known about children’s acceptance of social robots and its antecedents. To fill this research gap, this narrative review surveyed 34 articles on child-robot interaction published between 2000 and 2017. We focused on robot, user, and interaction characteristics as potential antecedents of children’s intentional and behavioral social robot acceptance. In general, children readily accept robots. However, we found that social, adaptive robot behavior, children’s sex and age, as well as frequency of the interaction seem to affect acceptance. Additionally, we found various theoretical and methodological shortcomings in the field. The review concludes with recommendations and directions for future research on children’s acceptance of social robots.

Evidence from empirical literature suggests that explainable complex behaviors can be built from structured compositions of explainable component behaviors with known properties. Such component behaviors can be built to directly perceive and exploit affordances. Using six examples of recent research in legged robot locomotion, we suggest that robots can be programmed to effectively exploit affordances without developing explicit internal models of them. We use a generative framework to discuss the examples, because it helps us to separate -- and thus clarify the relationship between -- description of affordance exploitation from description of the internal representations used by the robot in that exploitation. Under this framework, details of the architecture and environment are related to the emergent behavior of the system via a generative explanation. For example, the specific method of information processing a robot uses might be related to the affordance the robot is designed to exploit via a formal analysis of its control policy. By considering the mutuality of the agent-environment system during robot behavior design, roboticists can thus develop robust architectures which implicitly exploit affordances. The manner of this exploitation is made explicit by a well constructed generative explanation.

Appropriate robot behavior during human-robot interaction is a key part in the development of human-compliant assistive robotic systems. This study poses the question of how to continuously evaluate the quality of robotic behavior in a hybrid brain-computer interfacing (BCI) task, combining brain and non-brain signals, and how to use the collected information to adapt the robot's behavior accordingly. To this aim, we developed a rating system compatible with EEG recordings, requiring the users to execute only small movements with their thumb on a wireless controller to rate the robot's behavior on a continuous scale. The ratings were recorded together with dry EEG, respiration, ECG, and robotic joint angles in ROS. Pilot experiments were conducted with three users that had different levels of previous experience with robots. The results demonstrate the feasibility to obtain continuous rating data that give insight into the subjective user perception during direct human-robot interaction. The rating data suggests differences in subjective perception for users with no, moderate, or substantial previous robot experience. Furthermore, a variety of regression techniques, including deep CNNs, allowed us to predict the subjective ratings. Performance was better when using the position of the robotic hand than when using EEG, ECG, or respiration. A consistent advantage of features expected to be related to a motor bias could not be found. Across-user predictions showed that the models most likely learned a combination of general and individual features across-users. A transfer of pre-trained regressor to a new user was especially accurate in users with more experience. For future research, studies with more participants will be needed to evaluate the methodology for its use in practice. Data and code to reproduce this study are available at https://github.com/TNTLFreiburg/NiceBot.

In the last years, the growing interest in consensus and coordinated control of multi-agent mobile systems has originated the development of a variety of testbeds. These experimental platforms constitute an essential component in the research and education process in the robotics field. This paper describes the design and implementation of a testbed based on the position control and trajectory tracking of mini-sized ground robots. The testbed architecture consists of a monitoring system composed by a web cam and a control algorithm running in a main computer that communicates with the robots wirelessly. A unique feature of the testbed is the use of ArUco markers on top of the mobile platforms for motion tracking purposes using OpenCV and Visual Basic. Additionally, the testbed has a user graphic interface that allows to manage and visualize results of the robot behavior for the tracking of different pre-defined trajectories.

As robots are taking a more prominent role in our daily lives, it becomes increasingly important to consider how their presence influences us. Several studies have investigated effects of robot behavior on the extent to which that robot is positively evaluated. Likewise, studies have shown that the emotions a robot shows tend to be contagious: a happy robot makes us feel happy as well. It is unknown, however, whether the affect that people experience while interacting with a robot also influences their evaluation of the robot. This study aims to discover whether people's affective and evaluative responses to a social robot are related. Results show that affective responses and evaluations are related, and that these effects are strongest when a robot shows meaningful motions. These results are consistent with earlier findings in terms of how people evaluate social robots.

In this paper is described a four-wheeled autonomous robot vehicle that has as main application the feed of the animals (domestic or savage) found within an animal shelter, without the need for human presence. There are described both the hardware design of the robot (microcontroller, sensors, prime mover) and the artificial intelligence behind the robot behavior.

Most robots consistently repeat their motion with- out changes in a precise and consistent manner. But nowadays there are also robots able to dynamically change their motion and plan according to the people and environment that surround them. Furthermore, they are able to interact with humans and cooperate with them. With no information about the robot targets and intentions, the user feels uncomfortable even with a safe robot. In close human-robot collaboration, it is very important to make the user able to understand the robot intentions in a quick and intuitive way. In this work we have developed a system to use augmented reality to project directly into the workspace useful information. The robot intuitively shows its planned motion and task state. The AR module interacts with a vision system in order to display the changes in the workspace in a dynamic way. The representation of information about possible collisions and changes of plan allows the human to have a more comfortable and efficient interaction with the robot. The system is evaluated in different setups.

Assistive robotic arms enable users with physical disabilities to perform everyday tasks without relying on a caregiver. Unfortunately, the very dexterity that makes these arms useful also makes them challenging to teleoperate: the robot has more degrees-of-freedom than the human can directly coordinate with a handheld joystick. Our insight is that we can make assistive robots easier for humans to control by leveraging latent actions. Latent actions provide a low-dimensional embedding of high-dimensional robot behavior: for example, one latent dimension might guide the assistive arm along a pouring motion. In this paper, we design a teleoperation algorithm for assistive robots that learns latent actions from task demonstrations. We formulate the controllability, consistency, and scaling properties that user-friendly latent actions should have, and evaluate how different low-dimensional embeddings capture these properties. Finally, we conduct two user studies on a robotic arm to compare our latent action approach to both state-of-the-art shared autonomy baselines and a teleoperation strategy currently used by assistive arms. Participants completed assistive eating and cooking tasks more efficiently when leveraging our latent actions, and also subjectively reported that latent actions made the task easier to perform. The video accompanying this paper can be found at: https://youtu.be/wjnhrzugBj4.

We consider problems in which robots conspire to present a view of the world that differs from reality. The inquiry is motivated by the problem of validating robot behavior physically despite there being a discrepancy between the robots we have at hand and those we wish to study, or the environment for testing that is available versus that which is desired, or other potential mismatches in this vein. After formulating the concept of a convincing illusion, essentially a notion of system simulation that takes place in the real world, we examine the implications of this type of simulability in terms of infrastructure requirements. Time is one important resource: some robots may be able to simulate some others but, perhaps, only at a rate that is slower than real-time. This difference gives a way of relating the simulating and the simulated systems in a form that is relative. We establish some theorems, including one with the flavor of an impossibility result, and providing several examples throughout. Finally, we present data from a simple multi-robot experiment based on this theory, with a robot navigating amid an unbounded field of obstacles."Truth is beautiful, without doubt; but so are lies."—Ralph Waldo Emerson

Human-robot cooperation is increasingly demanded in industrial applications. Many tasks require the robot to enhance the capabilities of humans. In this scenario, safety also plays an important role in avoiding any accident involving humans, robots, and the environment. With this aim, the paper proposes a cooperative fuzzy-impedance control with embedded safety rules to assist human operators in heavy industrial applications while manipulating unknown weight parts. The proposed methodology is composed by four main components: (i) an inner Cartesian impedance controller (to achieve the compliant robot behavior), (ii) an outer fuzzy controller (to provide the assistance to the human operator), (iii) embedded safety rules (to limit force/velocity during the human-robot interaction enhancing safety), and (iv) a neural network approach (to optimize the control parameters for the human-robot collaboration on the basis of the target indexes of assistance performance defined for this purpose). The main achieved result refers to the capability of the controller to deal with uncertain payloads while assisting and empowering the human operator, both embedding in the controller safety features at force and velocity levels and minimizing the proposed performance indexes. The effectiveness of the proposed approach is verified with a KUKA iiwa 14 R820 manipulator in an experimental procedure where human subjects evaluate the robot performance in a collaborative lifting task of a 10 kg part.

The article presents approaches to the development of robotic systems based on the use of virtual prototypes, known as digital twins. Also, usage of such systems for walking robotic mechanisms are demonstrated by the example of developing an experimental debugging platform that supports both the design of algorithms for robot behavior at the upper level of the control hierarchy and the design of control algorithms for its nodes at the lower level. Authors used different basic tooling systems integrated with each other through cross-platform tools, such as Gazebo, V-Rep, IAR Embedded Workbench, Quartus, Fuzzy51 and several others.

Human interactions are affected by society's norms and cultural criteria. Individuals expect certain behaviors from other people, who have an obligation to act according to the expected behavior in the cultural and social context that surrounds them. Giving robots the ability to interact with humans, in human terms, is a great open challenge. People are willing to accept robotic systems in everyday life when robots engage in socially desirable (correct) behaviors with benevolent interaction styles (safe physical interaction). In addition, allowing robots to reason in social situations, involves a set of social norms that generate expectations, which can improve the dynamics of interactions between humans and robots, and the processes of selfassessment of robot behavior. In this article, we present a set of requirements for a more secure and socially correct physical interaction the robot with people, which is dependent on cultural criteria. Universidad de Costa Rica/[834-B7-267]/UCR/Costa Rica UCR::Vicerrectoría de Docencia::Ingeniería::Facultad de Ingeniería::Escuela de Ciencias de la Computación e Informática UCR::Vicerrectoría de Investigación::Unidades de Investigación::Ingeniería::Centro de Investigaciones en Tecnologías de Información y Comunicación (CITIC)

Sensitive manipulators have become wide-spread in collaborative robot applications and the industry is aiming for ways to use the advantages of Human-Robot-Collaboration (HRC) in an efficient way. Nevertheless most applications do not go beyond the coexistence of humans and robots. To facilitate a closer collaboration between humans and robots we examine the ability of sensitive robots to safeguard contact situations with humans. More precisely, we study the relation between certain robot parameters and the resulting biomechanical load to humans during a potential collision. As use case we consider a generic abstraction of a pick and place application in a laboratory setup. Particularly, the effects of changing the end effector velocity and varying the pick and place position were investigated. To do so, potential contact situations within the range of the robots workspace are evaluated. A statistical model of the robot behavior is presented for the investigated application scenario, and the experimental findings are elaborated in a theoretical context. It is discussed how the deduced information can be utilized to safeguard a collaborative robot application with a flexible pick and place position and how this holistic approach leads to a safe and efficient integration of collaborative tasks in general.

The usage of robot is rapidly growth in our society. The communication link and applications connect the robots to their clients or users. This communication link and applications are normally connected through some kind of network connections. This network system is amenable of being attached and vulnerable to the security threats. It is a critical part for ensuring security and privacy for robotic platforms. The paper, also discusses about several cyber-physical security threats that are only for robotic platforms. The peer to peer applications use in the robotic platforms for threats target integrity, availability and confidential security purposes. A Remote Administration Tool (RAT) was introduced for specific security attacks. An impact oriented process was performed for analyzing the assessment outcomes of the attacks. Tests and experiments of attacks were performed in simulation environment which was based on Gazbo Turtlebot simulator and physically on the robot. A software tool was used for simulating, debugging and experimenting on ROS platform. Integrity attacks performed for modifying commands and manipulated the robot behavior. Availability attacks were affected for Denial-of-Service (DoS) and the robot was not listened to Turtlebot commands. Integrity and availability attacks resulted sensitive information on the robot.

The curse of dimensionality is a challenge in many applications of Linear Temporal Logic robot motion planning, and is linked to the discretization of the robot's workspace. The discretization aggravates the problem by introducing a multitude of atomic propositions. This paper argues that a large portion of these atomic propositions is unnecessary. It demonstrates this point by introducing local navigation functions within a temporal logic planning framework, and utilizing register automata for reactive motion planning without explicit, high-resolution workspace discretization. Motivation for this approach comes from applications in pediatric motor rehabilitation involving play-based social child-robot interactions, where the appropriate robot behavior in response to child actions is best described in temporal logic terms. A simulation example drawn from the aforementioned pediatric rehabilitation studies illustrates the advantages of the approach.

The ability to provide comprehensive explanations of chosen actions is a hallmark of intelligence. Lack of this ability impedes the general acceptance of AI and robot systems in critical tasks. This paper examines what forms of explanations best foster human trust in machines and proposes a framework in which explanations are generated from both functional and mechanistic perspectives. The robot system learns from human demonstrations to open medicine bottles using (i) an embodied haptic prediction model to extract knowledge from sensory feedback, (ii) a stochastic grammar model induced to capture the compositional structure of a multistep task, and (iii) an improved Earley parsing algorithm to jointly leverage both the haptic and grammar models. The robot system not only shows the ability to learn from human demonstrators but also succeeds in opening new, unseen bottles. Using different forms of explanations generated by the robot system, we conducted a psychological experiment to examine what forms of explanations best foster human trust in the robot. We found that comprehensive and real-time visualizations of the robot’s internal decisions were more effective in promoting human trust than explanations based on summary text descriptions. In addition, forms of explanation that are best suited to foster trust do not necessarily correspond to the model components contributing to the best task performance. This divergence shows a need for the robotics community to integrate model components to enhance both task execution and human trust in machines.

To improve the design of a person-following robot, this preliminary study evaluates the influence of user tasks on human preferences of the robot's following angle and human perceptions of the robot's behavior. 32 participants were followed by a robot at three different following angles twice, once with an auditory task and once with a visual task, for a total of six walking trials. Results indicate that the type of user task influences participant preferences and perceptions. For the auditory task, as the following angle increased, participants were more satisfied with the robot's following behavior. For the visual task, as the following angle increased, participants were less satisfied with the robot's following behavior. In addition, participants were more perceptive of the robot's following behavior for the auditory task compared to the visual task. Additional research is required to better understand whether human preferences and perceptions depend on task modality or task complexity.

When performing tasks in uncertain environments and around humans, robots are likely to collide unexpectedly with people or objects. In order to ensure safety, most approaches rely on collision avoidance and try to prevent any contact from happening, which may result in unnecessary interruption of a task that would be feasible in spite of the obstacle. On the one hand, when an unexpected contact occurs, a safe robot behavior is required. On the other hand, it might be interesting to exploit the contact instead of moving away from it. In this paper, we present a contact-driven approach for safe and interactive robot operation to react to unforeseen contact events. This approach offers the possibility to control the contact while minimizing its effects on the robot tasks. It relies exclusively on the robot model and proprioceptive sensors. It is tested in simulation and hardware experiments on a 7 degrees of freedom robot arm and shows a safe contact behavior that does not interfere with the task, and as little as possible with the robot posture requirements.

In this paper, a force control is adopted to help an operator manipulate a robotic arm directly and efficiently in a path planning task. To achieve this, we mount a 6-DOF force sensor on the tip of the robotic arm. The operator can manipulate the robotic arm easily by pushing and pulling the force sensor. However, the main problem faced in this work is when the robot is pulled to a target by the operator, its end-effector will contact with surroundings with the result that the contact force repels the end-effector from the target. Thus, the robot attempts to attain the target position since there is an error between its current and target positions. Previous studies have shown that conservative control can suppress the rebounding effect, but maintaining it will cause the operator to suffer excessive resistance in subsequent operations. To solve this problem, we propose a collision rate to detect any collision by analyzing force behavior. Then, we use this rate to adjust the control gain of robot behavior. The proposed method was validated on an industrial robotic arm. The result shows that the robot produced a small resistance for manual manipulation and avoided bouncing behavior when touching any surroundings.

This live demonstration presents an audio-guided neuromorphic robot: from a Neuromorphic Auditory Sensor (NAS) to locomotion using Spiking Central Pattern Generators (sCPGs). Several gaits are generated by sCPGs implemented on a SpiNNaker board. The output of these sCPGs is sent in a real-time manner to an Field Programmable Gate Array (FPGA) board using an AER-to-SpiNN interface. The control of the hexapod robot joints is performed by the FPGA board. The robot behavior can be changed in real-time by means of the NAS. The audio information is sent to the SpiNNaker board which classifies it using a Spiking Neural Network (SNN). Thus, the input sound will activate a specific gait pattern which will eventually modify the behavior of the robot. Ministerio de Economía y Competitividad TEC2016-77785-P

Contains fulltext : 212483.pdf (Publisher’s version ) (Closed access) Robot-assisted therapy (RAT) offers potential advantages for improving the social skills of children with autism spectrum disorders (ASDs). This article provides an overview of the developed technology and clinical results of the EC-FP7-funded Development of Robot-Enhanced therapy for children with AutisM spectrum disorders (DREAM) project, which aims to develop the next level of RAT in both clinical and technological perspectives, commonly referred to as robot-enhanced therapy (RET). Within this project, a supervised autonomous robotic system is collaboratively developed by an interdisciplinary consortium including psychotherapists, cognitive scientists, roboticists, computer scientists, and ethicists, which allows robot control to exceed classical remote control methods, e.g., Wizard of Oz (WoZ), while ensuring safe and ethical robot behavior. Rigorous clinical studies are conducted to validate the efficacy of RET. Current results indicate that RET can obtain an equivalent performance compared to that of human standard therapy for children with ASDs. We also discuss the next steps of developing RET robotic systems. 10 p.

Orchard unmanned vehicle is a kind of agricultural robot. Aiming at the ultimate operation goal of orchard unmanned vehicle ‐‐ autonomous obstacle avoidance and navigation, the control system and the obstacle avoidance strategy based on traffic rules and robot behavior were proposed in this research. The control system structure of orchard unmanned vehicle system including orchard unmanned vehicle body, DSP development board, information acquisition module, data processing and navigation module and motion execution module was designed; the traffic rules for orchard unmanned vehicle were established, and the obstacle avoidance strategy based on behavior law was proposed, which mainly includes three sub-behaviors: Move-to-Goal, Avoid-Obstacle and Random-Noise; according to the actual working environment of orchard unmanned vehicle, the collision avoidance planning and obstacle avoidance process design were finally carried out. The research could provide theoretical basis for the obstacle avoidance simulation and practical test of unmanned vehicle in orchard.

Classification of complex acoustic scenes under real time scenarios is an active domain which has engaged several researchers lately form the machine learning community. A variety of techniques have been proposed for acoustic patterns or scene classification including natural soundscapes such as rain/thunder, and urban soundscapes such as restaurants/streets, etc. In this work, we present a framework for automatic acoustic classification for behavioral robotics. Motivated by several texture classification algorithms used in computer vision, a modified feature descriptor for sound is proposed which incorporates a combination of 1-D local ternary patterns (1D-LTP) and baseline method Mel-frequency cepstral coefficients (MFCC). The extracted feature vector is later classified using a multi-class support vector machine (SVM), which is selected as a base classifier. The proposed method is validated on two standard benchmark datasets i.e., DCASE and RWCP and achieves accuracies of 97.38 % and 94.10 % , respectively. A comparative analysis demonstrates that the proposed scheme performs exceptionally well compared to other feature descriptors.

One of the advantages of teaching robots by demonstration is that it can be more intuitive for users to demonstrate rather than describe the desired robot behavior. However, when the human demonstrates the task through an interface, the training data may inadvertently acquire artifacts unique to the interface, not the desired execution of the task. Being able to use one's own body usually leads to more natural demonstrations, but those examples can be more difficult to translate to robot control policies. This paper quantifies the benefits of using a virtual reality system that allows human demonstrators to use their own body to perform complex manipulation tasks. We show that our system generates superior demonstrations for a deep neural network without introducing a correspondence problem. The effectiveness of this approach is validated by comparing the learned policy to that of a policy learned from data collected via a conventional gaming system, where the user views the environment on a monitor screen, using a Sony Play Station 3 (PS3) DualShock 3 wireless controller as input.

State-of-the-art social robot navigation algorithms often lack a thorough experimental validation in human environments: simulated evaluations are often conducted under unrealistically strong assumptions that prohibit deployment in real world environments; experimental demonstrations that are limited in sample size do not provide adequate evidence regarding the user experience and the robot behavior; field studies may suffer from the noise imposed by uncontrollable factors from the environment; controlled lab experiments often fail to properly enforce challenging interaction settings. This paper contributes a first step towards addressing the outlined gaps in the literature. We present an original experiment, designed to test the implicit interaction between a mobile robot and a group of navigating human participants, under challenging settings in a controlled lab environment. We conducted a large-scale, within-subjects design study with 105 participants, exposed to three different conditions, corresponding to three distinct navigation strategies, executed by a telepresence robot (two autonomous, one teleoperated). We analyzed observed human and robot trajectories, under close interaction settings and participants' impressions regarding the robot's behavior. Key findings, extracted from a comparative statistical analysis include: (1) evidence that human acceleration is lower when navigating around an autonomous robot compared to a teleoperated one; (2) the lack of evidence to support the conventional expectation that teleoperation would be humans' preferred strategy. To the best of our knowledge, our study is unique in terms of goals, settings, thoroughness of evaluation and sample size.

It has long been assumed that when people observe robots they intuitively ascribe mind and intentionality to them, just as they do to humans. However, much of this evidence relies on experimenter-provided questions or self-reported judgments. We propose a new way of investigating people's mental state ascriptions to robots by carefully studying explanations of robot behavior. Since people's explanations of human behavior are deeply grounded in assumptions of mind and intentional agency, explanations of robot behavior can reveal whether such assumptions similarly apply to robots. We designed stimulus behaviors that were representative of a variety of robots in diverse contexts and ensured that people saw the behaviors as equally intentional, desirable, and surprising across both human and robot agents. We provided 121 participants with verbal descriptions of these behaviors and asked them to explain in their own words why the agent (human or robot) had performed them. To systematically analyze the verbal data, we used a theoretically grounded classification method to identify core explanation types. We found that people use the same conceptual toolbox of behavior explanations for both human and robot agents, robustly indicating inferences of intentionality and mind. But people applied specific explanatory tools at somewhat different rates and in somewhat different ways for robots, revealing specific expectations people hold when explaining robot behaviors.

Recently, deep neural networks (DNNs) are actively used for object recognition, but also for action control, so that an autonomous system, such as the robot, can perform human-like behaviors and operations. Unlike recognition tasks, real-time operation is important in action control, and it is too slow to use remote learning on a server communicating through a network. New learning techniques, such as reinforcement learning (RL), are needed to determine and select the correct robot behavior locally. Fig. 7.4.1(a) shows an example of a robot agent that uses a pre-trained DNN without RL, and Fig. 7.4.1(b) depicts an autonomous robot agent that learns continuously in the environment using RL. The agent without RL falls down if the land slope changes, but the RL-based agent iteratively collects walking experiences and learns to walk even though the land slope changes.