This paper details recent progress in modelling and understanding the processing, acquisition and evolution of grounded spatial language. It summarises key insights from the project and gives an overview of achievements possible today in integrated AI research programs.

We provide a systematic analysis of levels of integration between discrete high-level reasoning and continuous low-level reasoning to address hybrid planning problems in robotics. We identify four distinct strategies for such an integration: (i) low-level checks are done for all possible cases in advance and then this information is used during plan generation, (ii) low-level checks are done exactly when they are needed during the search for a plan, (iii) first all plans are computed and then infeasible ones are filtered, and (iv) by means of replanning, after finding a plan, low-level checks identify whether it is infeasible or not; if it is infeasible, a new plan is computed considering the results of previous low-level checks. We perform experiments on hybrid planning problems in robotic manipulation and legged locomotion domains considering these four methods of integration, as well as some of their combinations. We analyze the usefulness of levels of integration in these domains, both from the point of view of computational efficiency (in time and space) and from the point of view of plan quality relative to its feasibility. We discuss advantages and disadvantages of each strategy in the light of experimental results and provide some guidelines on choosing proper strategies for a given domain.

While there has been recent successes in learning single control policies, there are several several open challenges in robot motor skill learning. Firstly, many motor tasks can be solved in multiple ways, and, hence, we need to be able to learn each of these solutions as separate options from which the agent can choose from. Furthermore, we need to learn how to adapt an option to the current situation. Finally, we need to be able to combine several options sequentially in order to solve an overall-task. As we want to use our method for real robots, a high data efficiency is a natural additional requirement for motor skill learning. In this paper we summarize our work on information-theoretic motor skill learning. We show how to adapt the relative entropy policy search (REPS) algorithm for learning parametrized options and extend the algorithm in a mathematical sound way such that it can meet all these requirements. Finally, we summarize our experiments conducted on real robots.

The ability to specify a task without having to write special software is an important and prominent feature for a mobile service robot deployed in a crowded office environment, working around and interacting with people. In this paper, we contribute an interactive approach for enabling the users to teach tasks to a mobile service robot through verbal commands. The input is given as typed or spoken instructions, which are then mapped to the available sensing and actuation primitives on the robot. The main contributions of this work are the addition of conditionals on sensory information that the specified actions to be executed in a closed-loop manner, and a correction mode that allows an existing task to be modified or corrected at a later time by providing a replacement action during the test execution. We describe all the components of the system along with the implementation details and illustrative examples in depth. We also discuss the extensibility of the presented system, and point out potential future extensions.

As robotic technologies mature, we can imagine an increasing number of applications in which robots could soon prove to be useful in unstructured human environments. Many of those applications require a natural interface between the robot and untrained human users or are possible only in a human-robot collaborative scenario. In this paper, we study an example of such scenario in which a visually impaired person and a robotic guide collaborate in an unfamiliar environment. We then analyze how the scenario can be realized through language- and gesture-based human-robot interaction, combined with semantic spatial understanding and reasoning, and propose an integration of semantic world model with language and gesture models for several collaboration modes. We believe that this way practical robotic applications can be achieved in human environments with the use of currently available technology.

We propose the use of causality-based formal representation and automated reasoning methods to endow multiple teams of robots in a factory, with high-level cognitive capabilities, such as, optimal planning and diagnostic reasoning. We introduce algorithms for finding optimal decoupled plans and diagnosing the cause of a failure/discrepancy (e.g., robots may get broken or tasks may get reassigned to teams). We discuss how these algorithms can be embedded in an execution and monitoring framework, and show their applicability on an intelligent painting factory scenario.

In this paper we are considering an autonomous robot moving purposefully in a crowd of people (a marketplace). The robot should take into consideration the social costs of its movement, expressed in terms of violation of the personal space of the humans, blocking their path or even making physical contact with them. On the other hand, the full avoidance of any social cost might jeopardize the mission of the robot - in a sufficiently dense crowd, movement is impossible without violating at least some social norms. The individuals in the crowd, including the robot, periodically encounter micro-conflicts where they need to change their behavior to avoid large social costs (such as bumping into each other). We model the resolution of micro-conflicts as a series of games where the payoffs integrate the social and mission costs of the action choices. We show that human behavior corresponds to a strategy which is not necessarily optimal on a single-game basis; instead, it reflects the personality and social status of the person and the psychological requirement of consistency in behavior. We describe a robot behavior which classifies the strategy used by the opponent in the micro-conflict and chooses an appropriate counter-strategy which takes into account the urgency of the robot's mission.

Adverse drug events are a leading cause of danger and cost in health care. We could reduce both the danger and the cost if we had accurate models to predict, at prescription time for each drug, which patients are most at risk for known adverse reactions to that drug, such as myocardial infarction (MI, or "heart attack") if given a Cox2 inhibitor, angioedema if given an ACE inhibitor, or bleeding if given an anticoagulant such as Warfarin. We address this task for the specific case of Cox2 inhibitors, a type of non-steroidal anti-inflammatory drug (NSAID) or pain reliever that is easier on the gastrointestinal system than most NSAIDS. Because of the MI adverse drug reaction, some but not all very effective Cox2 inhibitors were removed from the market. Specifically, we use machine learning to predict which patients on a Cox2 inhibitor would suffer an MI. An important issue for machine learning is that we do not know which of these patients might have suffered an MI even without the drug. To begin to make some headway on this important problem, we compare our predictive model for MI for patients on Cox2 inhibitors against a more general model for predicting MI among a broader population not on Cox2 inhibitors.

A short discussion of the Spewy Louie Jr. No-Limit and Limit Texas Hold’em poker bot is presented. The hand clustering algorithm and the non-traditional game tree data- structure used are discussed in detail.

When solving extensive-form games with large action spaces, typically significant abstraction is needed to make the problem manageable from a modeling or computational perspective. When this occurs, a procedure is needed to interpret actions of the opponent that fall outside of our abstraction (by mapping them to actions in our abstraction). This is called an action translation mapping. Prior action translation mappings have been based on heuristics without theoretical justification. We show that the prior mappings are highly exploitable and that most of them violate certain natural desiderata. We present a new mapping that satisfies these desiderata and has significantly lower exploitability than the prior mappings. Furthermore, we observe that the cost of this worst-case performance benefit (low exploitability) is not high in practice; our mapping performs competitively with the prior mappings against no-limit Texas Hold'em agents submitted to the 2012 Annual Computer Poker Competition. We also observe several paradoxes that can arise when performing action abstraction and translation; for example, we show that it is possible to improve performance by including suboptimal actions in our abstraction and excluding optimal actions.