Think Fast: Real-Time Kinodynamic Belief-Space Planning for Projectile Interception

2025-11-30

AI Summary

Real-time interception of high-speed moving targets under sensor noise remains challenging due to tight temporal constraints and state uncertainty. Method: This paper proposes a tree-based motion planning framework integrating dynamic belief updating. It employs RIG-AKF—a robust, adaptive Kalman filter—for online estimation of the target’s state distribution; constructs a reachability tree in state-time space using dynamical primitives; and enables belief-evolution-driven value re-evaluation and seamless multi-target switching. The framework supports closed-loop, active decision-making and smooth trajectory re-planning. Contribution/Results: To our knowledge, this is the first work embedding uncertainty-aware belief updating into a dynamical tree structure, significantly enhancing responsiveness and robustness under stringent time constraints. Experiments on an ABB IRB-1600 manipulator equipped with a ZED 2i stereo camera demonstrate stable interception of high-speed projectiles, achieving an average response latency of <50 ms—validating both real-time performance and engineering feasibility.

Technology Category

• Intelligent Robots: State Estimation
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Intercepting fast moving objects, by its very nature, is challenging because of its tight time constraints. This problem becomes further complicated in the presence of sensor noise because noisy sensors provide, at best, incomplete information, which results in a distribution over target states to be intercepted. Since time is of the essence, to hit the target, the planner must begin directing the interceptor, in this case a robot arm, while still receiving information. We introduce an tree-like structure, which is grown using kinodynamic motion primitives in state-time space. This tree-like structure encodes reachability to multiple goals from a single origin, while enabling real-time value updates as the target belief evolves and seamless transitions between goals. We evaluate our framework on an interception task on a 6 DOF industrial arm (ABB IRB-1600) with an onboard stereo camera (ZED 2i). A robust Innovation-based Adaptive Estimation Adaptive Kalman Filter (RIAE-AKF) is used to track the target and perform belief updates.

Problem

• Real-time planning for intercepting fast-moving objects
• Handling sensor noise and incomplete target state information
• Dynamic belief updates during interception with robot arm

Innovation

• Tree structure with kinodynamic motion primitives
• Real-time belief updates for target interception
• Robust adaptive Kalman filter for tracking