Patrick D. Ulam

Moral Decision Making in Autonomous Systems: Enforcement, Moral Emotions, Dignity, Trust, and Deception

As humans are being progressively pushed further downstream in the decision-making process of autonomous systems, the need arises to ensure that moral standards, however defined, are adhered to by these robotic artifacts. While meaningful inroads have been made in this area regarding the use of ethical lethal military robots, including work by our laboratory, these needs transcend the warfighting domain and are pervasive, extending to eldercare, robot nannies, and other forms of service and entertainment robotic platforms. This paper presents an overview of the spectrum and specter of ethical issues raised by the advent of these systems, and various technical results obtained to date by our research group, geared towards managing ethical behavior in autonomous robots in relation to humanity. This includes: 1) the use of an ethical governor capable of restricting robotic behavior to predefined social norms; 2) an ethical adaptor which draws upon the moral emotions to allow a system to constructively and proactively modify its behavior based on the consequences of its actions; 3) the development of models of robotic trust in humans and its dual, deception, drawing on psychological models of interdependence theory; and 4) concluding with an approach towards the maintenance of dignity in human-robot relationships.

Integrated Mission Specification and Task Allocation for Robot Teams - Design and Implementation

As the capabilities, range of missions, and the size of robot teams increase, the ability for a human operator to account for all the factors in these complex scenarios can become exceedingly difficult. Our previous research has studied the use of case-based reasoning (CBR) tools to assist a user in the generation of multi-robot missions. These tools, however, typically assume that the robots available for the mission are of the same type (i.e., homogeneous). We loosen this assumption through the integration of contract-net protocol (CNP) based task allocation coupled with a CBR-based mission specification wizard. Two alternative designs are explored for combining case-based mission specification and CNP-based team allocation as well as the tradeoffs that result from the selection of one of these approaches over the other.

An ethical adaptor: Behavioral modification derived from moral emotions

This paper presents the motivation, basis and a prototype implementation of an ethical adaptor capable of using a moral affective function, guilt, as a basis for altering a robots ongoing behavior. While the research is illustrated in the context of the battlefield, the methods described are believed generalizable to other domains such as eldercare and are potentially extensible to a broader class of moral emotions, including compassion and empathy.

Using Optimal Foraging Models to Evaluate Learned Robotic Foraging Behavior

A key challenge in designing robot teams is determining how to allocate team members to specific roles according to their abilities and the demands of the environment. In this paper we explore this issue in the context of multi-robot foraging, and we show that optimal foraging theory can be used to evaluate our work in learned multi-robot foraging tasks. We present a means by which members of a multi-robot team may use reinforcement learning to allocate themselves to specific foraging roles appropriate to their environment and their abilities. We test this approach in environments with different distributions of various types of attractors and by varying the relative effectiveness of different foraging strategies. We then examine the effectiveness of the algorithm by comparing the distributions learned by the individual robots to those predicted by several optimal foraging models. We show the resulting learned distributions are substantially similar to those predicted by the optimal foraging theory models.

Multi-robot User Interface Modeling

This paper investigates the problem of user interface design and evaluation for autonomous teams of heterogeneous mobile robots. We explore an operator modeling approach to multi-robot user interface evaluation. Specifically the authors generated GOMS models, a type of user model, to investigate potential interface problems and to guide the interface development process. Results indicate that our interface design changes improve the usability of multi-robot mission generation substantially. We conclude that modeling techniques such as GOMS can play an important role in robotic interface development. Moreover, this research indicates that these techniques can be performed in an inexpensive and timely manner, potentially reducing the need for costly and demanding usability studies.

Lek behavior as a model for multi-robot systems

Lek behavior is a biological mechanism used by male birds to attract mates by forming a group. This project explores the use of a biological behavior found in many species of birds to form leks to guide the creation of groups of robots. The lek behavior provides a sound basis for multi-robot formation because it demonstrates a group of individual entities forming up around a scarce resource. This behavior can be useful to robots in many situations, with an example scenario the case in which robots were dropped via parachute into an area and then needed to form meaningful task-oriented groups.

When good communication go bad: communications recovery for multi-robot teams

Ad-hoc networks among groups of autonomous mobile robots are becoming a common occurrence as teams of robots take on increasingly complicated missions over wider areas. Research has often focused on proactive means in which the individual robots of the team may prevent communication failures between nodes in this network. This is not always possible especially in unknown or hostile environments. This research addresses reactive aspects of communication recovery. How should the members of the team react in the event of unseen communication failures between some or all of the nodes in the network? We present a number of behaviors to be utilized in the event of communications failure as well as a behavioral sequencer to further enhance the effectiveness of these recovery behaviors. The performance of the communication recovery behavior is analyzed in simulation and their application on hardware platforms is discussed.

Mission Specification and Control for Unmanned Aerial and Ground Vehicles for Indoor Target Discovery and Tracking

This paper describes ongoing research by Georgia Tech into the challenges of tasking and controlling heterogonous teams of unmanned vehicles in mixed indoor/outdoor reconnaissance scenarios. We outline the tools and techniques necessary for an operator to specify, execute, and monitor such missions. The mission specification framework used for the purposes of intelligence gathering during mission execution are first demonstrated in simulations involving a team of a single autonomous rotorcraft and three ground-based robotic platforms. Preliminary results including robotic hardware in the loop are also provided.

Biasing behavioral activation with intent for an entertainment robot

Deliberate control of an entertainment robot presents a special problem in balancing the requirement for intentional behavior with the existing mechanisms for autonomous action selection. It is proposed that the intentional biasing of activation in lower-level reactive behaviors is the proper mechanism for realizing such deliberative action. In addition, it is suggested that directed intentional bias can result in goal-oriented behavior without subsuming the underlying action selection used to generate natural behavior. This objective is realized through a structure called the intentional bus. The intentional bus serves as the interface between deliberative and reactive control by realizing high-level goals through the modulation of intentional signals sent to the reactive layer. A deliberative architecture that uses the intentional bus to realize planned behavior is described. In addition, it is shown how the intentional bus framework can be expanded to support the serialization of planned behavior by shifting from direct intentional influence for plan execution to attentional triggering of a learned action sequence. Finally, an implementation of this architecture, developed and tested on Sony’s humanoid robot QRIO, is described.