Researchers have introduced an adaptive path planning framework that significantly reduces computational search effort and shortens trajectory lengths for autonomous robotic systems operating in cluttered spaces. Developed to navigate geometrically constrained obstacles, the proposed algorithm, designated as AFI-RRT*, integrates global guidance mechanisms with localized free-space exploration to optimize autonomous motion. Experimental evaluations demonstrate that the technique consistently achieves superior path quality while lowering overall computational burdens compared to standard sampling-based planning algorithms.
Autonomous path planning remains a persistent technical challenge for mobile ground robots and unmanned aerial vehicles, particularly in cluttered environments where systems must achieve efficient exploration, short trajectory generation, and minimal processing overhead. Traditional Rapidly-exploring Random Tree Star, or RRT*, frameworks along with their various modifications can produce high-quality navigation trajectories. However, these conventional approaches frequently suffer from excessive space exploration and generate unnecessarily large search trees when navigating through geometrically constrained settings.
To overcome these structural limitations, researchers Amir Seyyedabbasi, Erdal Alimovski, and Burak Aggul developed the Adaptive Free Interval guided RRT* architecture. The framework combines coarse A*-based global guidance with structured local free-space exploration. This architectural design focuses sampling routines directly through adaptive free-interval layers, allowing the planner to concentrate search efforts specifically on geometrically relevant free-space regions while still preserving necessary global exploration capabilities throughout the surrounding domain.
The operational architecture of AFI-RRT* provides several key computational advantages over standard random exploration techniques. By focusing sampling through structured layers, the algorithm substantially curtails unnecessary tree expansion into unpromising regions of the environment. Furthermore, the framework enables the direct extraction of the final path solution straight from the generated search tree, entirely eliminating the requirement for secondary geometric post-processing or auxiliary trajectory smoothing steps.
The research team comprehensively evaluated the performance of AFI-RRT* across nine distinct experimental environments, executing thirty independent test runs per condition to ensure statistical rigor. The study benchmarked the novel method against seven established navigation algorithms: standard RRT*, Informed RRT*, original RRT, RRT-Connect, QS-RRT*, AdaRRT, and the deterministic A* search method. Across all experimental settings, the adaptive framework maintained an uninterrupted one hundred percent success rate in identifying valid navigation paths.
In comparative benchmark assessments, AFI-RRT* recorded the lowest overall mean path length among the evaluated planning algorithms. When measured directly against standard RRT*, the proposed method decreased average execution time by 13.8 percent and reduced average search-tree size by 52.5 percent. These experimental measurements confirm that focusing sampling on structured intervals prevents the unchecked proliferation of exploratory nodes within complex navigation maps.
The efficiency gains remained pronounced when comparing the novel framework against Informed RRT*, yielding reductions of 35.7 percent in average execution time and 53.0 percent in search-tree size. Statistical evaluations via Wilcoxon signed-rank tests confirmed that path-length improvements achieved by AFI-RRT* were statistically significant across all nine test environments relative to standard RRT*, and across seven environments relative to Informed RRT*, validating the architecture’s balance of trajectory quality and computational economy.
Subject of Research: Adaptive sampling-based path planning for mobile robots and unmanned aerial vehicles
Article Title: AFI-RRT\(^*\): Adaptive Free Interval Guided RRT\(^*\) for Path Planning in Cluttered Environments
Article References: Seyyedabbasi, A., Alimovski, E., & Aggul, B. (2026). AFI-RRT$$^*$$: Adaptive Free Interval Guided RRT$$^*$$ for Path Planning in Cluttered Environments. International Journal of Computational Intelligence Systems. https://doi.org/10.1007/s44196-026-01674-5
Image Credits: AI Generated
DOI: 10.1007/s44196-026-01674-5
Keywords: path planning, AFI-RRT*, RRT*, robotics, mobile robots, unmanned aerial vehicles, cluttered environments, adaptive free interval, A* algorithm, Informed RRT*, search tree size, computational efficiency, execution time, sampling-based planning, Wilcoxon signed-rank test, path length, free space exploration
Cite Scienmag News
Denise Maddox. (November 18, 2026). AFI-RRT* Framework Optimizes Robot Path Planning in Cluttered Environments. Scienmag. https://scienmag.com/afi-rrt-framework-optimizes-robot-path-planning-in-cluttered-environments/
Denise Maddox. "AFI-RRT* Framework Optimizes Robot Path Planning in Cluttered Environments." Scienmag, 18 November 2026, https://scienmag.com/afi-rrt-framework-optimizes-robot-path-planning-in-cluttered-environments/. Accessed 18 November 2026.
Denise Maddox. "AFI-RRT* Framework Optimizes Robot Path Planning in Cluttered Environments." Scienmag. November 18, 2026. https://scienmag.com/afi-rrt-framework-optimizes-robot-path-planning-in-cluttered-environments/








