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.
more »
« less
Decentralized Multi-agent information-theoretic control for target estimation and localization: finding gas leaks
This article presents a new decentralized multi-agent information-theoretic (DeMAIT) control algorithm for mobile sensors (agents). The algorithm leverages Bayesian estimation and information-theoretic motion planning for efficient and effective estimation and localization of a target, such as a chemical gas leak. The algorithm consists of: (1) a non-parametric Bayesian estimator, (2) an information-theoretic trajectory planner that generates “informative trajectories” for agents to follow, and (3) a controller and collision avoidance algorithm to ensure that each agent follows its trajectory as closely as possible in a safe manner. Advances include the use of a new information-gain metric and its analytical gradient, which do not depend on an infinite series like prior information metrics. Dynamic programming and multi-threading techniques are applied to efficiently compute the mutual information to minimize measurement uncertainty. The estimation and motion planning processes also take into account the dynamics of the sensors and agents. Extensive simulations are conducted to compare the performance between the DeMAIT algorithm to a traditional raster-scanning method and a clustering method with coordination. The main hypothesis that the DeMAIT algorithm outperforms the other two methods is validated, specifically where the average localization success rate for the DeMAIT algorithm is (a) higher and (b) more robust to changes in the source location, robot team size, and search area size than the raster-scanning and clustering methods. Finally, outdoor field experiments are conducted using a team of custom-built aerial robots equipped with gas concentration sensors to demonstrate efficacy of the DeMAIT algorithm to estimate and find the source of a propane gas leak.
more » « less- PAR ID:
- 10547616
- Publisher / Repository:
- SAGE Publications
- Date Published:
- Journal Name:
- The International Journal of Robotics Research
- Volume:
- 39
- Issue:
- 13
- ISSN:
- 0278-3649
- Format(s):
- Medium: X Size: p. 1525-1548
- Size(s):
- p. 1525-1548
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
We present an incremental scalable motion planning algorithm for finding maximally informative trajectories for decentralized mobile robots. These robots are deployed to observe an unknown spatial field, where the informativeness of observations is specified as a density function. Existing works that are typically restricted to discrete domains and synchronous planning often scale poorly depending on the size of the problem. Our goal is to design a distributed control law in continuous domains and an asynchronous communication strategy to guide a team of cooperative robots to visit the most informative locations within a limited mission duration. Our proposed Asynchronous Information Gathering with Bayesian Optimization (AsyncIGBO) algorithm extends ideas from asynchronous Bayesian Optimization (BO) to efficiently sample from a density function. It then combines them with decentralized reactive motion planning techniques to achieve efficient multi-robot information gathering activities. We provide a theoretical justification for our algorithm by deriving an asymptotic no-regret analysis with respect to a known spatial field. Our proposed algorithm is extensively validated through simulation and real-world experiment results with multiple robots.more » « less
-
more » « less
This article focuses on enabling an aerial robot to fly through multiple openings at high speed using image-based estimation, planning, and control. State-of-the-art approaches assume that the robot’s global translational variables (e.g., position and velocity) can either be measured directly with external localization sensors or estimated onboard. Unfortunately, estimating the translational variables may be impractical because modeling errors and sensor noise can lead to poor performance. Furthermore, monocular-camera-based pose estimation techniques typically require a model of the gap (window) in order to handle the unknown scale. Herein, a new scheme for image-based estimation, aggressive-maneuvering trajectory generation, and motion control is developed for multi-rotor aerial robots. The approach described does not rely on measurement of the translational variables and does not require the model of the gap or window. First, the robot dynamics are expressed in terms of the image features that are invariant to rotation (invariant features). This step decouples the robot’s attitude and keeps the invariant features in the flat output space of the differentially flat system. Second, an optimal trajectory is efficiently generated in real time to obtain the dynamically-feasible trajectory for the invariant features. Finally, a controller is designed to enable real-time, image-based tracking of the trajectory. The performance of the estimation, planning, and control scheme is validated in simulations and through 80 successful experimental trials. Results show the ability to successfully fly through two narrow openings, where the estimation and planning computation and motion control from one opening to the next are performed in real time on the robot.
-
more » « less
Abstract Collaboration requires agents to coordinate their behavior on the fly, sometimes cooperating to solve a single task together and other times dividing it up into sub‐tasks to work on in parallel. Underlying the human ability to collaborate is theory‐of‐mind (ToM), the ability to infer the hidden mental states that drive others to act. Here, we develop Bayesian Delegation, a decentralized multi‐agent learning mechanism with these abilities. Bayesian Delegation enables agents to rapidly infer the hidden intentions of others by inverse planning. We test Bayesian Delegation in a suite of multi‐agent Markov decision processes inspired by cooking problems. On these tasks, agents with Bayesian Delegation coordinate both their high‐level plans (e.g., what sub‐task they should work on) and their low‐level actions (e.g., avoiding getting in each other's way). When matched with partners that act using the same algorithm, Bayesian Delegation outperforms alternatives. Bayesian Delegation is also a capable ad hoc collaborator and successfully coordinates with other agent types even in the absence of prior experience. Finally, in a behavioral experiment, we show that Bayesian Delegation makes inferences similar to human observers about the intent of others. Together, these results argue for the centrality of ToM for successful decentralized multi‐agent collaboration.
-
We present a closed-loop multi-arm motion planner that is scalable and flexible with team size. Traditional multi-arm robotic systems have relied on centralized motion planners, whose run times often scale exponentially with team size, and thus, fail to handle dynamic environments with open-loop control. In this paper, we tackle this problem with multi-agent reinforcement learning, where a shared policy network is trained to control each individual robot arm to reach its target end-effector pose given observations of its workspace state and target end-effector pose. The policy is trained using Soft Actor-Critic with expert demonstrations from a sampling-based motion planning algorithm (i.e., BiRRT). By leveraging classical planning algorithms, we can improve the learning efficiency of the reinforcement learning algorithm while retaining the fast inference time of neural networks. The resulting policy scales sub-linearly and can be deployed on multi-arm systems with variable team sizes. Thanks to the closed-loop and decentralized formulation, our approach generalizes to 5-10 multiarm systems and dynamic moving targets (>90% success rate for a 10-arm system), despite being trained on only 1-4 arm planning tasks with static targets.more » « less
