Scenario-Based Trajectory Optimization in Uncertain Dynamic Environments

Groot, Oscar de; Brito, Bruno; Ferranti, Laura; Gavrila, Dariu M.; Alonso–Mora, Javier

doi:10.1109/lra.2021.3074866

Cited by 19 publications

(25 citation statements)

References 57 publications

(158 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While this paper focuses on the robot's own performance, our approach may be adapted to account for social coordination and altruism [30] in cooperative human-robot settings. Finally, there is an open opportunity to derive formal probabilistic or worst-case safety guarantees based on results established in [16,18]. The Laplace approximation used by (8) requires the mean function (human's rational action) µ M i (x t , u R t ) as the maximizer of Q M i x t , u R t , u M i .…”

Section: Discussion and Future Workmentioning

confidence: 99%

See 1 more Smart Citation

Active Uncertainty Reduction for Human-Robot Interaction: An Implicit Dual Control Approach

Hu¹,

Fisac²

2022

Preprint

View full text Add to dashboard Cite

Predictive models are effective in reasoning about human motion, a crucial part that affects safety and efficiency in human-robot interaction. However, robots often lack access to certain key parameters of such models, for example, human's objectives, their level of distraction, and willingness to cooperate. Dual control theory addresses this challenge by treating unknown parameters as stochastic hidden states and identifying their values using information gathered during control of the robot. Despite its ability to optimally and automatically trade off exploration and exploitation, dual control is computationally intractable for general human-in-the-loop motion planning, mainly due to nested trajectory optimization and human intent prediction. In this paper, we present a novel algorithmic approach to enable active uncertainty learning for human-in-the-loop motion planning based on the implicit dual control paradigm. Our approach relies on sampling-based approximation of stochastic dynamic programming, leading to a model predictive control problem that can be readily solved by real-time gradient-based optimization methods. The resulting policy is shown to preserve the dual control effect for generic human predictive models with both continuous and categorical uncertainty. The efficacy of our approach is demonstrated with simulated driving examples.

show abstract

Section: Discussion and Future Workmentioning

confidence: 99%

“…P [x ∈ F] ≤ β, where β is the tolerance level, provided that x ∈ F can be written as a set of inequality constraints. An analytical bound of β and probabilistic feasibility guarantee are established for chance-constrained ST-SMPC problems with dynamic obstacles in [16]. -Soft constrained.…”

Section: Active Uncertainty Learning For Human-robot Interactionmentioning

confidence: 99%

Active Uncertainty Reduction for Human-Robot Interaction: An Implicit Dual Control Approach

Hu¹,

Fisac²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…MPC is a popular technique for real-time collision avoidance for autonomous driving [6]- [9] and robotics [10]- [14]. A typical MPC algorithm computes control inputs by solving SHN and FB are with the Model Predictive Control Laboratory, UC Berkeley.…”

Section: Related Workmentioning

confidence: 99%

“…2) Process Noise Distribution: We model [w t n t ] as a product of 6 (2 for agent, 2×2 for obstacles) independent uni-variate truncated normal random variables (cf. [9], [10]), truncNorm(µ, σ, a, b), with µ = 0, a = −2, b = 2 and σ = 0.01 for w t , σ = 0.1 for n t . The resulting distribution for [w t n t ] has mean 0, variance Σ = 7.7 • blkdiag(10 −5 I 2×2 , 10 −3 I 4×4 ) and support D = ([−0.02, 0.02]) 2 × ([−0.2, 0.2]) 4 (with Γ = blkdiag(10 2 I2×2, 10I4×4), γ = 2 ), which is used for defining the uncertainty descriptions D1, D2, D3.…”

Section: A Longitudinal Control Example 1) Models and Geometrymentioning

confidence: 99%

Collision Avoidance for Dynamic Obstacles with Uncertain Predictions using Model Predictive Control

Nair¹,

Tseng²,

Borrelli³

2022

Preprint

View full text Add to dashboard Cite

We propose a Model Predictive Control (MPC) for collision avoidance between an autonomous agent and dynamic obstacles with uncertain predictions. The collision avoidance constraints are imposed by enforcing positive distance between convex sets representing the agent and the obstacles, and tractably reformulating them using Lagrange duality. This approach allows for smooth collision avoidance constraints even for polytopes, which otherwise require mixed-integer or nonsmooth constraints. We consider three widely used descriptions of the uncertain obstacle position: 1) Arbitrary distribution with polytopic support, 2) Gaussian distributions and 3) Arbitrary distribution with first two moments known. For each case we obtain deterministic reformulations of the collision avoidance constraints. The proposed MPC formulation optimizes over feedback policies to reduce conservatism in satisfying the collision avoidance constraints. The proposed approach is validated using simulations of traffic intersections in CARLA.

show abstract

“…The authors in [13] consider robust constraint satisfaction leading to highly conservative trajectories. More recent works such as [12], [18]- [20] leverage probability distribution of the uncertainty or past samples to guarantee that collision avoidance constraints are satisfied with high probability. Authors in [13], [21] focus primarily on autonomous driving scenarios.…”

Section: Introductionmentioning

confidence: 99%

Data-Driven Risk-sensitive Model Predictive Control for Safe Navigation in Multi-Robot Systems

Navsalkar¹,

Hota²

2022

Preprint

View full text Add to dashboard Cite

Safe navigation is a fundamental challenge in multi-robot systems due to the uncertainty surrounding the future trajectory of the robots that act as obstacles for each other. In this work, we propose a principled data-driven approach where each robot repeatedly solves a finite horizon optimization problem subject to collision avoidance constraints with latter being formulated as distributionally robust conditional value-atrisk (CVaR) of the distance between the agent and a polyhedral obstacle geometry. Specifically, the CVaR constraints are required to hold for all distributions that are close to the empirical distribution constructed from observed samples of prediction error collected during execution. The generality of the approach allows us to robustify against prediction errors that arise under commonly imposed assumptions in both distributed and decentralized settings. We derive tractable finite-dimensional approximations of this class of constraints by leveraging convex and minmax duality results for Wasserstein distributionally robust optimization problems. The effectiveness of the proposed approach is illustrated in a multi-drone navigation setting implemented in Gazebo platform.

show abstract

Scenario-Based Trajectory Optimization in Uncertain Dynamic Environments

Cited by 19 publications

References 57 publications

Active Uncertainty Reduction for Human-Robot Interaction: An Implicit Dual Control Approach

Active Uncertainty Reduction for Human-Robot Interaction: An Implicit Dual Control Approach

Collision Avoidance for Dynamic Obstacles with Uncertain Predictions using Model Predictive Control

Data-Driven Risk-sensitive Model Predictive Control for Safe Navigation in Multi-Robot Systems

Contact Info

Product

Resources

About