Answers are grounded in this paper and cite their sources.
Autonomous Industrial Vehicles (AIVs) operating in warehouses and industrial facilities must navigate constrained layouts while responding to stationary and moving obstacles.A path that is geometrically collision-free may still produce undesirable motion if vehicle dynamics and terminal behavior are not considered during control.This paper presents SteadyPath, a simulation-based framework for stability-aware autonomous industrial maneuvering.The framework combines a virtual perception layer based on simulator-generated ground-truth object positions and distances, vehicle-state estimation, obstacle-aware decision-making, A* path planning, and Model Predictive Control with terminal constraints.The MPC controller predicts future vehicle motion over a finite horizon and selects steering and longitudinal control actions while seeking a stable terminal condition.The system is evaluated using four scenario tests and a quantitative comparison against a baseline controller.In the reported mission benchmark, the proposed controller completed the mission in 76.4 s compared with 85.6 s for the baseline, reduced mean tracking error from 14.61 m to 2.52 m, reduced peak lateral deviation from 18.91 m to 5.25 m, and increased the clearance buffer margin from 1.30 m to 3.60 m.Both controllers recorded zero dynamic detours or re-plans in the benchmark.The proposed controller was recorded as reaching an invariant stable terminal set, while the baseline was recorded as having a lagging boundary.Scenario testing showed successful outcomes for normal-road, static-obstacle, and fully blocked-road cases; the moving-obstacle case requires clarification because its textual result says the obstacle was handled safely while its status is marked FAIL.The findings indicate improved performance on the reported simulation metrics, while further testing is required to establish robustness across broader dynamic conditions and real sensor inputs.
No takes yet. Share an insight, caveat, or question.
M et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: