Presentation is loading. Please wait.

Presentation is loading. Please wait.

The Purpose of HART (Hedonism, Autonomy, Responsibility, Trust) Niranjan Suri Maria Gini, Seyed Waqar Jaffry, Niranjan Suri, Janneke van der Zwaan, Arnoud.

Similar presentations


Presentation on theme: "The Purpose of HART (Hedonism, Autonomy, Responsibility, Trust) Niranjan Suri Maria Gini, Seyed Waqar Jaffry, Niranjan Suri, Janneke van der Zwaan, Arnoud."— Presentation transcript:

1 The Purpose of HART (Hedonism, Autonomy, Responsibility, Trust) Niranjan Suri Maria Gini, Seyed Waqar Jaffry, Niranjan Suri, Janneke van der Zwaan, Arnoud Visser

2 Long list issues (purpose related)  Killer applications (different team sizes)  What is teamwork?  Role of communication in teamwork (reduction inside good team)  Role of training inside team  How to communicate H->R, R->H, R->R.  Role of none-robot agents inside teams  Multiple representation (both HAR) in same role  How can RA be aware of the human and his feelings?  Metrics; teams, performance, meta-information about the experiments, quantitative and qualitative methodologies  Emotions for the robots to support the group interactions  Transparency / deceiving  Lifetime of a team (team formation / role taking / specialization / breaking up)  Training is partly done by the environment / partly from the team-members  Do robots still need a identity for scale > 100  Explore / control / autonomy -> gradual responsibility

3 Short list issues (purpose related)  Metrics; teams, performance, meta-information about the experiments, quantitative and qualitative methodologies 1, 2  How can RA be aware of the human and his feelings? 2, 2  Lifetime of a team (team formation / role taking / specialization / breaking up) 1, 3  Killer applications (different team sizes) 2, 3  How to communicate H->R, R->H, R->R. 1  Do robots still need a identity for scale > 100 1  Multiple representation (both HAR) in same role 3  Emotions for the robots to support the group interactions 3

4 Short list issues (purpose related)  Metrics; teams, performance, meta-information about the experiments, quantitative and qualitative methodologies 1, 2  How can RA be aware of the human and his feelings? 2, 2  Lifetime of a team (team formation / role taking / specialization / breaking up) 1, 3  Killer applications (different team sizes) -> Milestones  How to communicate H->R, R->H, R->R. 1  Do robots still need a identity for scale > 100 1  Multiple representation (both HAR) in same role 3  Emotions for the robots to support the group interactions 3

5 Milestone 1&2 (killer applications)  Example of team of 10 (2 drivers / 8 robots)  Convoy of trucks with two humans: easier on the highway than behind the warfront  Traffic lights should be intelligent before the convoy is allowed on highway  The army application should be operational in 2016  Ethics / laws will be the main issue to get this scenario operational before 2020 at the highways

6 Milestone 3 (killer applications)  Example of team of 2 (1 nurse / 1 robot)  Lifting assistance for a nurse  Alternatives:  Making the bed more intelligent (too restrictive)  Get the lifting capability in the environment (a lot of unfriendly infrastructure)  Exoskelet for the nurse (unfriendly for the other tasks of the nurse) Lifting robot is most flexible and friendly solution  The challenge of lifting is still at the mechanical department. Should we already work on the teamwork?  The robot should automatically follow the nurse and keep out of the nurse’s way  The robot should know when to follow and when to stay (judging the intentions)  Milestone at 2020

7 Milestone 4 (killer applications)  Example of team of 5 (2 operators / 1 human explorer / 1 ground robot / 1 air robot)  Urban Search and Rescue  Milestone January 2011

8

9 Trust  Hard to gain, easy to lose  How do robots gain and maintain human trust  Do robots need to trust humans?  Predictability of system is key to trust  Prescriptive notion – if the system is performing as specified by the objective, then it can be trusted  Another indicator – “normal” communication between team members  Effective conveyance of performance and limitations to humans  How does the human recognize and convey to the robot detrimental environmental conditions  Analogy to Coaching – need to understand the limits, why something is going wrong, and convey options / direction to the robots  Dynamics of Human Trust – key role in team building  Teaming is something that is perceived by humans, not the robots  Robots must adapt their behavior based on their perception of the human trust model  Robots must be able to interpret indirect human expression / communication

10 Autonomy and Responsibility  “A Robot without a battery is a very autonomous robot” – Visser  “A Robot that can say no is a very autonomous robot” - Gini  Meta question – what qualifies as a robot? Is Autonomy Required?  Bulldozer?, Robonaut? Jackhammer? Car?  Behavior-based Robotics – two goals:  Invariants that should not be violated (do not run into walls, kill anyone, including self, etc.)  Attainable goals – related to task or objective  How does teaching (e.g., by demonstration) / learning fit into this?  How does learning for a robot differ from learning for humans?  Awareness of the human team member(s) is paramount  To what extent, if at all, do we need to anthropomorphize robots? Not just from a physical perspective, but from the notions of trust, behavior, etc.  Responsibility – four types?  Responsibility for the given / assigned / delegated tasks  Responsibility to communicate with team members for shared SA  Responsibility to the other team members / task (e.g., help other team members)  Responsibility for the greater good (e.g., not polluting the environment, etc.)

11 Applications  Search and Rescue  UAV – Mountain Search (Brown University)  RoboCup City-Level Search and Rescue for Disaster Recovery  Persistent Surveillance  Fixed and mobile assets  Cooperative Medicine  Nurse’s aid  Operating room  Rehabilitation  Assistants for Elderly Citizens  Warfighting  Useful to categorize applications into good targets for teams of size 2, 5, and 10 members  Nurse’s aid, Elderly Citizen Assistants (2)  Mountain Search and Rescue, Persistent Surveillance (5)  Warfighting (10)

12 Metrics  What are appropriate metrics to measure success of teamwork?  We don’t want a Turing Test, but some test  One Approach is to show that Human + Robot can do more than Human + Human or Robot + Robot  Scalability with respect to numbers  What are the best domains to show HART?  Meta metric – how “much” teamwork is there in a solution?  Metric – how well is the solution working?  Is there a notion of a local (i.e., individual) objective or metric versus a global (i.e., systemwide) objective or metric?  How does robustness play into this? Should robustness be an independent metric?  “Types” of robustness – flexibility to deal with novel/different situations (opposite of brittleness)  Role substitution – humanoid robots substituting for humans  Fit – how well do robots fit in an environment constructed by and for humans

13 Questions  What is essential for teamwork?  Human-level communication (gestures, facial expressions)?


Download ppt "The Purpose of HART (Hedonism, Autonomy, Responsibility, Trust) Niranjan Suri Maria Gini, Seyed Waqar Jaffry, Niranjan Suri, Janneke van der Zwaan, Arnoud."

Similar presentations


Ads by Google