A5003213012- AI-based Problem Solving and Planning
- Multi-Agent Systems and Negotiation
- Logic, Reasoning, and Knowledge
- Semantic Web and Ontologies
- Business Process Modeling and Analysis
- Bayesian Modeling and Causal Inference
- Explainable Artificial Intelligence (XAI)
- Natural Language Processing Techniques
- Robotic Path Planning Algorithms
- Robot Manipulation and Learning
- Topic Modeling
- Reinforcement Learning in Robotics
- Human-Automation Interaction and Safety
- Robotics and Automated Systems
- Social Robot Interaction and HRI
- Ethics and Social Impacts of AI
- Robotic Process Automation Applications
- Virtual Reality Applications and Impacts
- Augmented Reality Applications
- Bacillus and Francisella bacterial research
- Adversarial Robustness in Machine Learning
- Anomaly Detection Techniques and Applications
- Speech and dialogue systems
- Machine Learning and Algorithms
- Teleoperation and Haptic Systems
Arizona State University
2014-2021
Human Computer Interaction (Switzerland)
2020
Cambridge Scientific (United States)
2017-2019
IBM (United States)
2019
Decision Systems (United States)
2017
Jadavpur University
2011-2013
When AI systems interact with humans in the loop, they are often called on to provide explanations for their plans and behavior. Past work plan primarily involved system explaining correctness of its rationale decision terms own model. Such soliloquy is wholly inadequate most realistic scenarios where have domain task models that differ significantly from used by system. We posit best studied light these differing models. In particular, we show how explanation can be seen as a "model...
Virtual, Augmented, and Mixed Reality for Human-Robot Interaction (VAM-HRI) has been gaining considerable attention in HRI research recent years. However, the community lacks a set of shared terminology framework characterizing aspects mixed reality interfaces, presenting serious problems future research. Therefore, it is important to have common terms concepts that can be used precisely describe organize diverse array work being done within field. In this article, we present novel taxonomic...
Intelligent robots and machines are becoming pervasive in human populated environments. A desirable capability of these agents is to respond goal-oriented commands by autonomously constructing task plans. However, such autonomy can add significant cognitive load potentially introduce safety risks humans when behave unexpected ways. Hence, for be helpful, one important requirement them synthesize plans that easily understood humans. While there exists previous work studied socially acceptable...
Recent work in explanation generation for decision making agents has looked at how unexplained behavior of autonomous systems can be understood terms differences the model system and human's understanding same, process as a result this mismatch then seen reconciliation these models. Existing algorithms such settings, while having been built on contrastive, selective social properties explanations studied extensively psychology literature, have not, to best our knowledge, evaluated settings...
There has been significant interest of late in generating behavior agents that is interpretable to the human (observer) loop. However, work this area typically lacked coherence on topic, with proposed solutions for “explicable”, “legible”, “predictable” and “transparent” planning overlapping, sometimes conflicting, semantics all aimed at some notion understanding what intentions observer will ascribe an agent by observing its behavior. This also true recent works “security” “privacy” plans...
The 1st International Workshop on Virtual, Augmented, and Mixed Reality for Human-Robot Interactions (VAM-HRI) will bring together HRI, Robotics, Artificial Intelligence, researchers to identify challenges in mixed reality interactions between humans robots. Topics relevant the workshop include development of robots that can interact with reality, use virtual developing interactive robots, design new augmented interfaces mediate communication comparisons capabilities perceptions agents, best...
In this paper, we provide a comprehensive outline of the different threads work in Explainable AI Planning (XAIP) that has emerged as focus area last couple years and contrast with earlier efforts field terms techniques, target users, delivery mechanisms. We hope survey will guidance to new researchers automated planning towards role explanations effective design human-in-the-loop systems, well established researcher some perspective on evolution exciting world explainable planning.
This is a demonstration of our newly released Python package NL2LTL which leverages the latest in natural language understanding (NLU) and large models (LLMs) to translate instructions linear temporal logic (LTL) formulas. allows direct translation formal languages that reasoning system can use, while at same time, allowing end-user provide inputs without having understand any details an underlying language. The comes with support for set default LTL patterns, corresponding popular DECLARE...
Among the many anticipated roles for robots in future is that of being a human teammate. Aside from all technological hurdles have to be overcome with respect hardware and control make fit work humans, added complication here humans conscious subconscious expectations their teammates - indeed, we argue teaming mostly cognitive rather than physical coordination activity. This introduces new challenges AI robotics community requires fundamental changes traditional approach design autonomy....
Beliefs play an important role in human-robot teaming scenarios, where the robots must reason about other agents' intentions and beliefs order to inform their own plan generation process, successfully coordinate plans with agents. In this paper, we cast evolving complex structure of beliefs, inference over them, as a planning recognition problem. We use agent modeled terms predicates create automated problem instance, which is then used along known complete domain model predict whose are...
Recent advances in mixed-reality technologies have renewed interest alternative modes of communication for human-robot interaction. However, most the work this direction has been confined to tasks such as teleoperation, simulation or explication individual actions a robot. In paper, we will discuss how capability project intentions affect task planning capabilities Specifically, start with discussion on projection can be used reveal information regarding future robot at time execution. We...
Model reconciliation has been proposed as a way for an agent to explain its decisions human who may have different understanding of the same planning problem by explaining in terms these model differences.However, often human's mental (and hence difference) is not known precisely and such explanations cannot be readily computed.In this paper, we show how explanation generation process evolves presence uncertainty or incompleteness generating {\em conformant explanations} that are applicable...
Recently there has been a lot of focus on human robot co-habitation issues that are often orthogonal to many aspects human-robot teaming; e.g. producing socially acceptable behaviors robots and de-conflicting plans humans in shared environments. However, an interesting offshoot these settings largely overlooked is the problem planning for serendipity - i.e. stigmergic collaboration without explicit commitments agents co-habitation. In this paper we formalize notion first time, provide...
It has long been assumed that for effective human-robot teaming, it is desirable assistive robots to infer the goals and intents of humans, take proactive actions help them achieve their goals. However, there not any systematic evaluation accuracy this claim. On face it, are several ways a robot assistant can in fact reduce effectiveness teaming. For example, increase cognitive load human teammate by performing unanticipated human. In such cases, even though teaming performance could be...
While natural language systems continue improving, they are still imperfect. If a user has better understanding of how system works, may be able to accomplish their goals even in imperfect systems. We explored whether explanations can support effective authoring utterances and those impact users' mental models the context that generates small programs. Through an online study (n=252), we compared two main types explanations: 1) system-focused, which provide information about processes...
Ambiguity and noise in natural language instructions create a significant barrier towards adopting autonomous systems into safety critical workflows involving humans machines. In this paper, we propose to build on recent advances electrophysiological monitoring methods augmented reality technologies, develop alternative modes of communication between robots involved large-scale proximal collaborative tasks. We will first introduce techniques for projecting robot's intentions its human...
Proactive Decision Support aims at improving the decision making experience of human decision-makers by enhancing quality decisions and ease them. Given that AI techniques are efficient in searching over a potentially large solution space (of decision) finding good solutions, it can be used for human-in-the-loop scenarios such as disaster response demand naturalistic making. A decision-maker, scenarios, may high-cognitive overload leading to loss situational awareness. In this paper, we...
In this paper we propose an energy efficient method for path planning by a robot arm. First have developed cost function that provides set of via points determining suitable according to the obstacles present in surroundings and other restrictions its motion. Then fit polynomial interior ensure journey is smooth takes place with consumption minimum energy. The via-points as well been determined using IWO (Invasive Weed Optimization) which new search heuristic based on colonizing property...