Legibility and predictability of robot motion

Dragan, Anca D.; Lee, Kenton C.T.; Srinivasa, Siddhartha S.

doi:10.1109/hri.2013.6483603

Cited by 466 publications

(356 citation statements)

References 35 publications

Supporting

Mentioning

348

Contrasting

Order By: Relevance

“…This cost, frequently used to encourage trajectory smoothness [29], produces trajectories that reach directly toward the goal, in line with users' expectation for a point robot (as evidenced in [13]). It also allows for an analytical V G and its gradient, making the optimization process very fast.…”

Section: B Parametersmentioning

confidence: 96%

“…Our previous work [13] formalized these notions and proposed mathematical models that measure how legible or predictable a motion is, which we summarize below. A main result is that the two properties are fundamentally different, and that a departure from predictability is often necessary to increase the legibility of the motion.…”

Section: Defining Legible Motionmentioning

confidence: 99%

“…In [13], we tested that a motion with higher Legibility score is indeed more legible to users, for both a point robot, as well as the robot in Fig.1.…”

Section: Defining Legible Motionmentioning

confidence: 99%

“…As our user studies show (Sec. VII, as well as our previous work [13]), some unpredictability is often necessary to convey intent -it is unpredictability beyond a threshold (like the outermost trajectory in Fig.1) that confuses users and lowers their confidence in what the robot is doing.…”

Section: Introductionmentioning

confidence: 99%

“…In our prior work ( [13], summarized in Sec. III), we proposed such a model based on the theory of action interpretation [12] in psychology, the result having strong motivations in the principle of rational action [17].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Generating Legible Motion

Dragan¹,

Srinivasa²

2013

Robotics: Science and Systems IX

109

119

View full text Add to dashboard Cite

Abstract-Legible motion -motion that communicates its intent to a human observer -is crucial for enabling seamless human-robot collaboration. In this paper, we propose a functional gradient optimization technique for autonomously generating legible motion. Our algorithm optimizes a legibility metric inspired by the psychology of action interpretation in humans, resulting in motion trajectories that purposefully deviate from what an observer would expect in order to better convey intent. A trust region constraint on the optimization ensures that the motion does not become too surprising or unpredictable to the observer.Our studies with novice users that evaluate the resulting trajectories support the applicability of our method and of such a trust region. They show that within the region, legibility as measured in practice does significantly increase. Outside of it, however, the trajectory becomes confusing and the users' confidence in knowing the robot's intent significantly decreases.

show abstract

Section: B Parametersmentioning

confidence: 96%

Section: Defining Legible Motionmentioning

confidence: 99%

“…In [13], we tested that a motion with higher Legibility score is indeed more legible to users, for both a point robot, as well as the robot in Fig.1.…”

Section: Defining Legible Motionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Generating Legible Motion

Dragan¹,

Srinivasa²

2013

Robotics: Science and Systems IX

109

119

View full text Add to dashboard Cite

show abstract

Explaining robot policies

et al. 2021

View full text Add to dashboard Cite

In order to interact with a robot or make wise decisions about where and how to deploy it in the real world, humans need to have an accurate mental model of how the robot acts in different situations. We propose to improve users' mental model of a robot by showing them examples of how the robot behaves in informative scenarios. We explore this in two settings. First, we show that when there are many possible environment states, users can more quickly understand the robot's policy if they are shown critical states where taking a particular action is important. Second, we show that when there is a distribution shift between training and test environment distributions, then it is more effective to show exploratory states that the robot does not visit naturally.

show abstract

Effects of movement speed and predictability in human–robot collaboration

Koppenborg

Nickel

Naber

et al. 2017

Hum Ftrs & Erg Mfg Svc

104

View full text Add to dashboard Cite

Human–robot collaboration (HRC) is characterized by a spatiotemporal overlap between the workspaces of the human and the robot and has become a viable option in manufacturing and other industries. However, for companies considering employing HRC it remains unclear how best to configure such a setup, because empirical evidence on human factors requirements remains inconclusive. As robots execute movements at high levels of automation, they adapt their speed and movement path to situational demands. This study therefore experimentally investigated the effects of movement speed and path predictability of an industrial collaborating robot on the human operator. Participants completed tasks together with a robot in an industrial workplace simulated in virtual reality. A lower level of predictability was associated with a loss in task performance, while faster movements resulted in higher‐rated values for task load and anxiety, indicating demands on the operator exceeding the optimum. Implications for productivity and safety and possible advancements in HRC workplaces are discussed.

show abstract

Legibility and predictability of robot motion

Cited by 466 publications

References 35 publications

Generating Legible Motion

Generating Legible Motion

Explaining robot policies

Effects of movement speed and predictability in human–robot collaboration

Contact Info

Product

Resources

About