Interaction-aware Conformal Prediction for Crowd Navigation

2025-02-10

AI Summary

Modeling the interdependent uncertainties between robot planning and human motion in dense crowd navigation remains challenging, as conventional approaches rely on static or independent assumptions. Method: This paper introduces the first interaction-aware conformal prediction framework that explicitly models human motion uncertainty as a function of robot trajectory planning. The approach integrates model predictive control (MPC), a learned trajectory predictor, a differentiable human behavior simulator, and a conformal prediction module—enabling decision-dependent calibration set construction and dynamic uncertainty interval estimation. Results: Evaluated in multi-density crowd simulations, the method significantly improves navigation safety, social compliance, uncertainty calibration accuracy, and real-time performance. The algorithm is lightweight and computationally efficient; its open-source implementation has been validated for practical deployment feasibility.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Planning under Uncertainty

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

During crowd navigation, robot motion plan needs to consider human motion uncertainty, and the human motion uncertainty is dependent on the robot motion plan. We introduce Interaction-aware Conformal Prediction (ICP) to alternate uncertainty-aware robot motion planning and decision-dependent human motion uncertainty quantification. ICP is composed of a trajectory predictor to predict human trajectories, a model predictive controller to plan robot motion with confidence interval radii added for probabilistic safety, a human simulator to collect human trajectory calibration dataset conditioned on the planned robot motion, and a conformal prediction module to quantify trajectory prediction error on the decision-dependent calibration dataset. Crowd navigation simulation experiments show that ICP strikes a good balance of performance among navigation efficiency, social awareness, and uncertainty quantification compared to previous works. ICP generalizes well to navigation tasks under various crowd densities. The fast runtime and efficient memory usage make ICP practical for real-world applications. Code is available at https://github.com/tedhuang96/icp.

Problem

• Addresses human motion uncertainty in robot navigation.
• Integrates decision-dependent uncertainty quantification and planning.
• Balances efficiency, social awareness, and safety in crowd navigation.

Innovation

• Interaction-aware Conformal Prediction
• Decision-dependent uncertainty quantification
• Probabilistic safety motion planning