Legibility and predictability of robot motion

By Anca Dragan, Kenton C. T. Lee, Siddhartha S. Srinivasa. 2013

800 citations recorded by the source corpus.

Subfields: Control and Systems Engineering, Social Psychology, Human-Computer Interaction, Robotics & Control

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Control and Systems Engineering

In Social Psychology

In Human-Computer Interaction

In Robotics & Control

Citations in the corpus

Learning Trajectory Preferences for Manipulators via Iterative Improvement (2013, background): “easons for different tasks. These methods are complementary to ours, in that we could leverage their database of trajectories and train our system on samples drawn from it. Other recent works such as [10, 32] consider generating human-like trajectories. These works are complementary to ours in that ”
The mythos of model interpretability (2016, background): “To contextualize any definition of interpretability, we first consider the motives that it addresses (expanded in §2).”
Machine Theory of Mind (2018, background): “They will also be highly useful for communication and pedagogy (Dragan et al., 2013; Fisac et al., 2017; Milli et al., 2017), and will thus likely play a key role in human-machine interaction.”
Do explanations make VQA models more predictable to a human? (2018, background): “Dragan et al. (2013) describe the intent-expressiveness of a robot as its trajectory being expressive of its goal.”
Generating Legible Motion (2013, methodology): “Our measurement method follows [13, 18]: we showed the users a video of the trajectory, and asked them to stop the video as soon as they felt confident in their prediction of which goal the robot is headed toward (Fig.”
Learning with Humans in the Loop (2020, methodology): “ON THE UTILITY OF MODEL LEARNING IN HUMAN-ROBOT INTERACTION 8 reinforcement learning, and the ToM-based method will plan the robot’s actions using the learned model.”
Modeling Others using Oneself in Multi-Agent Reinforcement Learning (2018, background): “Such models might be useful in human-robot or teacher-student interactions (Dragan et al., 2013; Fisac et al., 2017), as well as for value alignment problems (Hadﬁeld-Menell et al., 2016).”
Communicating Robot Arm Motion Intent Through Mixed Reality Head-mounted Displays (2017, background): “d animated storytelling techniques, such as forming suggestive poses or generating initial movements [37]. This increases legibility: the ability to infer the robot’s goal through its directed motion [10]. However, these methods still lack the ability to transparently communicate complex paths and m”
Implicit Communication of Actionable Information in Human-AI teams (2019, background): “If a person can correctly predict an action that somebody will take to achieve a known goal, we would say that the action is predictable; if a person can infer the goal based on her observation of an action, then we say that the action is legible [25].”
Core Challenges of Social Robot Navigation: A Survey (2021, background): “[56] defined legibility as the property of motion that enables an observer to quickly and confidently predict the goal of an actor.”
Progress and prospects of the human–robot collaboration (2017, positive): “…and anticipation (i.e., the property of predicting the human intention, the goal of the collaboration, and optimizing the robot control policy to take into account the human action, reducing the human effort or improving some shared performance criteria) [Stulp et al., 2015,Dragan et al., 2013].”
HumanPlus: Humanoid Shadowing and Imitation from Humans (2024, background): “Prior works develop humanoid and dexterous teleoperation by us-ing human motion capture suits [14, 17, 20, 21], ex-oskeletons [35, 35, 63, 70, 75], haptic feedback devices [6, 61, 71], and VR devices for visual feed-backs [7, 38, 59, 90] and for end-effector control [2, 46, 62, 86].”
Balancing Explicability and Explanation in Human-Aware Planning (2017, background): “ork in this direction has ranged on a spectrum of “human-aware planning” [2,3,8,11–13,22,33]wherearobotpassivelytriestoaccountforthe plans of humans cohabiting its workspace, to “explicable planning” [14, 23, 34, 35] where a robot generates plans that are explicable or predictable to a human observe”
A survey of robot learning from demonstrations for Human-Robot Collaboration (2017, background): “Anca D. Dragan et al. focused on the difference between two types of robot motions (predictable robot motion and legible motion) [11].”
Reward-rational (implicit) choice: A unifying formalism for reward learning (2020, background): “Boltzmann-rational policies are widespread in psychology [13, 14, 15], economics [16, 17, 18, 17], and AI [19, 20, 21, 22, 23] as models of human choices, actions, or inferences.”
Learning Rewards from Linguistic Feedback (2020, background): “This idea has been extended to settings in which agents are provided with intentionally informative demonstrations (Ho et al. 2016), a variety of human acts (Jeon, Milli, and Dragan 2020), or themselves act informatively (Dragan, Lee, and Srinivasa 2013; Hadﬁeld-Menell et al. 2016).”
Explainable Agents and Robots: Results from a Systematic Literature Review (2019, background): “Note that in the literature, and within the domain of social robots in particular, other related terms are also in use, including understandability [47], explicability2 [16], transparency [105], predictability [26], readability [98], and legibility [64].”