Andrea Bajcsy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Andrea Bajcsy has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Anca Diana Dragan
University of California, Berkeley, Berkeley, CA, USA
4 shared publications - Claire J. Tomlin
University of California, Berkeley
3 shared publications - Dylan P. Losey
Virginia Tech, Blacksburg, Virginia, USA
2 shared publications - Ellis Ratner
Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA
2 shared publications - Jaime Fernandez Fisac
Department of Electrical and Computer Engineering, Princeton University, Princeton, NJ, USA
2 shared publications - Anand Siththaranjan
Stanford University
1 shared publication
Latest Publications
- 17 November 2026
- Published by SAGE Publications in The International Journal of Robotics Research
A unified framework for robots that influence humans over long-term interaction
Robot actions influence the decisions of nearby humans. Here, influence refers to intentional change: robots influence humans when they shift the human’s behavior in a way that helps the robot complete its task. Imagine an autonomous car trying to merge; by proactively nudging into the human’s lane, the robot causes human drivers to yield and provide space. Influence is often necessary for seamless interaction. However, if influence is left unregulated and uncontrolled, robots will negatively impact the humans around them (e.g., autonomous cars that repeatedly merge in front of humans may cause human drivers to become more aggressive). Prior works have begun to address this problem by creating a variety of control algorithms that seek to influence humans. Although these methods are effective in the short-term, they fail to maintain influence over time as the human adapts to the robot’s behaviors. In this paper we therefore present an optimization framework that enables robots to purposely regulate their influence over humans across both short - term and long - term interactions. Here, the robot maintains its influence by reasoning over a dynamic human model which captures how the robot’s current choices will impact the human’s future behavior. Our resulting framework serves to unify current approaches: we demonstrate that state-of-the-art methods are simplifications of our underlying formalism. Our framework also provides a principled way to generate influential policies: in the best case the robot exactly solves our framework to find optimal, influential behavior. But when solving this optimization problem becomes impractical, designers can introduce their own simplifications to reach tractable approximations. We experimentally compare our unified framework to state-of-the-art baselines and ablations, and demonstrate across simulations and user studies that this framework is able to successfully influence humans over repeated interactions. See videos of our experiments here: https://youtu.be/nPekTUfUEbo .
- 17 November 2026
- Published by SAGE Publications in The International Journal of Robotics Research
A unified framework for robots that influence humans over long-term interaction
Robot actions influence the decisions of nearby humans. Here, influence refers to intentional change: robots influence humans when they shift the human’s behavior in a way that helps the robot complete its task. Imagine an autonomous car trying to merge; by proactively nudging into the human’s lane, the robot causes human drivers to yield and provide space. Influence is often necessary for seamless interaction. However, if influence is left unregulated and uncontrolled, robots will negatively impact the humans around them (e.g., autonomous cars that repeatedly merge in front of humans may cause human drivers to become more aggressive). Prior works have begun to address this problem by creating a variety of control algorithms that seek to influence humans. Although these methods are effective in the short-term, they fail to maintain influence over time as the human adapts to the robot’s behaviors. In this paper we therefore present an optimization framework that enables robots to purposely regulate their influence over humans across both short - term and long - term interactions. Here, the robot maintains its influence by reasoning over a dynamic human model which captures how the robot’s current choices will impact the human’s future behavior. Our resulting framework serves to unify current approaches: we demonstrate that state-of-the-art methods are simplifications of our underlying formalism. Our framework also provides a principled way to generate influential policies: in the best case the robot exactly solves our framework to find optimal, influential behavior. But when solving this optimization problem becomes impractical, designers can introduce their own simplifications to reach tractable approximations. We experimentally compare our unified framework to state-of-the-art baselines and ablations, and demonstrate across simulations and user studies that this framework is able to successfully influence humans over repeated interactions. See videos of our experiments here: https://youtu.be/nPekTUfUEbo .
- Top Cited Papers
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (3) , 2208-2215
Contingency Games for Multi-Agent Interaction
Contingency planning, wherein an agent generates a set of possible plans conditioned on the outcome of an uncertain event, is an increasingly popular way for robots to act under uncertainty. In this work we take a game-theoretic perspective on contingency planning, tailored to multi-agent scenarios in which a robot's actions impact the decisions of other agents and vice versa. The resulting contingency game allows the robot to efficiently interact with other agents by generating strategic motion plans conditioned on multiple possible intents for other actors in the scene. Contingency games are parameterized via a scalar variable which represents a future time when intent uncertainty will be resolved. By estimating this parameter online, we construct a game-theoretic motion planner that adapts to changing beliefs while anticipating future certainty. We show that existing variants of game-theoretic planning under uncertainty are readily obtained as special cases of contingency games. Through a series of simulated autonomous driving scenarios, we demonstrate that contingency games close the gap between certainty-equivalent games that commit to a single hypothesis and non-contingent multi-hypothesis games that do not account for future uncertainty reduction.
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (3) , 2208-2215
Top Cited PapersContingency Games for Multi-Agent Interaction
Contingency planning, wherein an agent generates a set of possible plans conditioned on the outcome of an uncertain event, is an increasingly popular way for robots to act under uncertainty. In this work we take a game-theoretic perspective on contingency planning, tailored to multi-agent scenarios in which a robot's actions impact the decisions of other agents and vice versa. The resulting contingency game allows the robot to efficiently interact with other agents by generating strategic motion plans conditioned on multiple possible intents for other actors in the scene. Contingency games are parameterized via a scalar variable which represents a future time when intent uncertainty will be resolved. By estimating this parameter online, we construct a game-theoretic motion planner that adapts to changing beliefs while anticipating future certainty. We show that existing variants of game-theoretic planning under uncertainty are readily obtained as special cases of contingency games. Through a series of simulated autonomous driving scenarios, we demonstrate that contingency games close the gap between certainty-equivalent games that commit to a single hypothesis and non-contingent multi-hypothesis games that do not account for future uncertainty reduction.
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (3) , 2303-2310
StROL: Stabilized and Robust Online Learning From Humans
Robots often need to learn the human's reward function online, during the current interaction. This real-time learning requires fast but approximate learning rules: when the human's behavior is noisy or suboptimal, current approximations can result in unstable robot learning. Accordingly, in this letter we seek to enhance the robustness and convergence properties of gradient descent learning rules when inferring the human's reward parameters. We model the robot's learning algorithm as a dynamical system over the human preference parameters, where the human's true (but unknown) preferences are the equilibrium point. This enables us to perform Lyapunov stability analysis to derive the conditions under which the robot's learning dynamics converge. Our proposed algorithm (StROL) uses these conditions to learn robust-by-design learning rules: given the original learning dynamics, StROL outputs a modified learning rule that now converges to the human's true parameters under a larger set of human inputs. In practice, these autonomously generated learning rules can correctly infer what the human is trying to convey, even when the human is noisy, biased, and suboptimal. Across simulations and a user study we find that StROL results in a more accurate estimate and less regret than state-of-the-art approaches for online reward learning.
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (3) , 2303-2310
StROL: Stabilized and Robust Online Learning From Humans
Robots often need to learn the human's reward function online, during the current interaction. This real-time learning requires fast but approximate learning rules: when the human's behavior is noisy or suboptimal, current approximations can result in unstable robot learning. Accordingly, in this letter we seek to enhance the robustness and convergence properties of gradient descent learning rules when inferring the human's reward parameters. We model the robot's learning algorithm as a dynamical system over the human preference parameters, where the human's true (but unknown) preferences are the equilibrium point. This enables us to perform Lyapunov stability analysis to derive the conditions under which the robot's learning dynamics converge. Our proposed algorithm (StROL) uses these conditions to learn robust-by-design learning rules: given the original learning dynamics, StROL outputs a modified learning rule that now converges to the human's true parameters under a larger set of human inputs. In practice, these autonomously generated learning rules can correctly infer what the human is trying to convey, even when the human is noisy, biased, and suboptimal. Across simulations and a user study we find that StROL results in a more accurate estimate and less regret than state-of-the-art approaches for online reward learning.
- Open Access
Deception Game: Closing the Safety-Learning Loop in Interactive Robot Autonomy
An outstanding challenge for the widespread deployment of robotic systems like autonomous vehicles is ensuring safe interaction with humans without sacrificing performance. Existing safety methods often neglect the robot's ability to learn and adapt at runtime, leading to overly conservative behavior. This paper proposes a new closed-loop paradigm for synthesizing safe control policies that explicitly account for the robot's evolving uncertainty and its ability to quickly respond to future scenarios as they arise, by jointly considering the physical dynamics and the robot's learning algorithm. We leverage adversarial reinforcement learning for tractable safety analysis under high-dimensional learning dynamics and demonstrate our framework's ability to work with both Bayesian belief propagation and implicit learning through large pre-trained neural trajectory predictors.
Open AccessDeception Game: Closing the Safety-Learning Loop in Interactive Robot Autonomy
An outstanding challenge for the widespread deployment of robotic systems like autonomous vehicles is ensuring safe interaction with humans without sacrificing performance. Existing safety methods often neglect the robot's ability to learn and adapt at runtime, leading to overly conservative behavior. This paper proposes a new closed-loop paradigm for synthesizing safe control policies that explicitly account for the robot's evolving uncertainty and its ability to quickly respond to future scenarios as they arise, by jointly considering the physical dynamics and the robot's learning algorithm. We leverage adversarial reinforcement learning for tractable safety analysis under high-dimensional learning dynamics and demonstrate our framework's ability to work with both Bayesian belief propagation and implicit learning through large pre-trained neural trajectory predictors.
- Open Access
- 13 March 2023
- Published by Association for Computing Machinery (ACM)
- p. 350-358
Towards Modeling and Influencing the Dynamics of Human Learning
- 13 March 2023
- Published by Association for Computing Machinery (ACM)
- p. 350-358
Open AccessTowards Modeling and Influencing the Dynamics of Human Learning
- 18 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 6 (2) , 2373-2380
Efficient Dynamics Estimation With Adaptive Model Sets
Robotic systems frequently operate under changing dynamics, such as driving across varying terrain, encountering sensing and actuation faults, or navigating around humans with uncertain and changing intent. In order to operate effectively in these situations, robots must be capable of efficiently estimating these changes in order to adapt at the decision-making, planning, and control levels. Typical estimation approaches maintain a fixed set of candidate models at each time step; however, this can be computationally expensive if the number of models is large. In contrast, we propose a novel algorithm that employs an adaptive model set. We leverage the idea that the current model set must be expanded if its models no longer sufficiently explain the sensor measurements. By maintaining only a small subset of models at each time step, our algorithm improves on efficiency; at the same time, by choosing the appropriate models to keep, we avoid compromising on performance. We show that our algorithm exhibits higher efficiency in comparison to several baselines, when tested on simulated manipulation, driving, and human motion prediction tasks, as well as in hardware experiments on a 7 DOF manipulator.
- 18 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 6 (2) , 2373-2380
Efficient Dynamics Estimation With Adaptive Model Sets
Robotic systems frequently operate under changing dynamics, such as driving across varying terrain, encountering sensing and actuation faults, or navigating around humans with uncertain and changing intent. In order to operate effectively in these situations, robots must be capable of efficiently estimating these changes in order to adapt at the decision-making, planning, and control levels. Typical estimation approaches maintain a fixed set of candidate models at each time step; however, this can be computationally expensive if the number of models is large. In contrast, we propose a novel algorithm that employs an adaptive model set. We leverage the idea that the current model set must be expanded if its models no longer sufficiently explain the sensor measurements. By maintaining only a small subset of models at each time step, our algorithm improves on efficiency; at the same time, by choosing the appropriate models to keep, we avoid compromising on performance. We show that our algorithm exhibits higher efficiency in comparison to several baselines, when tested on simulated manipulation, driving, and human motion prediction tasks, as well as in hardware experiments on a 7 DOF manipulator.
- 30 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 6 (1) , 24-31
A Robust Control Framework for Human Motion Prediction
Designing human motion predictors which preserve safety while maintaining robot efficiency is an increasingly important challenge for robots operating in close physical proximity to people. One approach is to use robust control predictors that safeguard against every possible future human state, leading to safe but often too conservative robot plans. Alternatively, intent-driven predictors explicitly model how humans make decisions given their intent, leading to efficient robot plans. However, when the intent model is misspecified, the robot might confidently plan unsafe maneuvers. In this letter, we combine ideas from robust control and intent-driven human modelling to formulate a novel human motion predictor which provides robustness against misspecified human models, but reduces the conservatism of traditional worst-case predictors. Our approach predicts the human states by trusting the intent-driven model to decide only which human actions are completely unlikely. We then safeguard against all likely enough actions, much like a robust control predictor. We demonstrate in simulation and hardware how our approach safeguards against misspecified human intent models while not leading to overly conservative robot plans.
- 30 September 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 6 (1) , 24-31
A Robust Control Framework for Human Motion Prediction
Designing human motion predictors which preserve safety while maintaining robot efficiency is an increasingly important challenge for robots operating in close physical proximity to people. One approach is to use robust control predictors that safeguard against every possible future human state, leading to safe but often too conservative robot plans. Alternatively, intent-driven predictors explicitly model how humans make decisions given their intent, leading to efficient robot plans. However, when the intent model is misspecified, the robot might confidently plan unsafe maneuvers. In this letter, we combine ideas from robust control and intent-driven human modelling to formulate a novel human motion predictor which provides robustness against misspecified human models, but reduces the conservatism of traditional worst-case predictors. Our approach predicts the human states by trusting the intent-driven model to decide only which human actions are completely unlikely. We then safeguard against all likely enough actions, much like a robust control predictor. We demonstrate in simulation and hardware how our approach safeguards against misspecified human intent models while not leading to overly conservative robot plans.
- 24 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Robotics
- Vol. 36 (3) , 835-854
Quantifying Hypothesis Space Misspecification in Learning From Human–Robot Demonstrations and Physical Corrections
The human input has enabled autonomous systems to improve their capabilities and achieve complex behaviors that are otherwise challenging to generate automatically. Recent work focuses on how robots can use such inputs—such as, demonstrations or corrections—to learn intended objectives. These techniques assume that the human‘s desired objective already exists within the robot’s hypothesis space. In reality, this assumption is often inaccurate: there will always be situations where the person might care about aspects of the task that the robot does not know about. Without this knowledge, the robot cannot infer the correct objective. Hence, when the robot’s hypothesis space is misspecified, even methods that keep track of uncertainty over the objective fail because they reason about which hypothesis might be correct, and not whether any of the hypotheses are correct. In this article, we posit that the robot should reason explicitly about how well it can explain human inputs given its hypothesis space and use that situational confidence to inform how it should incorporate the human input. We demonstrate our method on a 7 degrees-of-freedom robot manipulator in learning from two important types of human inputs: demonstrations of motion planning tasks and physical corrections during the robot’s task execution.
- 24 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Robotics
- Vol. 36 (3) , 835-854
Quantifying Hypothesis Space Misspecification in Learning From Human–Robot Demonstrations and Physical Corrections
The human input has enabled autonomous systems to improve their capabilities and achieve complex behaviors that are otherwise challenging to generate automatically. Recent work focuses on how robots can use such inputs—such as, demonstrations or corrections—to learn intended objectives. These techniques assume that the human‘s desired objective already exists within the robot’s hypothesis space. In reality, this assumption is often inaccurate: there will always be situations where the person might care about aspects of the task that the robot does not know about. Without this knowledge, the robot cannot infer the correct objective. Hence, when the robot’s hypothesis space is misspecified, even methods that keep track of uncertainty over the objective fail because they reason about which hypothesis might be correct, and not whether any of the hypotheses are correct. In this article, we posit that the robot should reason explicitly about how well it can explain human inputs given its hypothesis space and use that situational confidence to inform how it should incorporate the human input. We demonstrate our method on a 7 degrees-of-freedom robot manipulator in learning from two important types of human inputs: demonstrations of motion planning tasks and physical corrections during the robot’s task execution.