The Bulldozer Technique: Efficient Elimination of Local Minima Traps for APF-Based Robot Navigation

TL;DR

The paper addresses local minima in artificial potential field (APF) based robot navigation by introducing Bulldozer, a block-based field modification that backfills low-potential regions and, when needed, uses a ramp to escape traps. It preserves APF's simplicity and real-time performance while embedding a systematic local-minima elimination mechanism and refined path planning via within-block optimization and Dijkstra detours. Hardware experiments on a Pioneer P3DX demonstrate superior robustness and faster execution compared with standard APF, A*, and PRM baselines, with tracking validated by a Kanayama-based controller that yields smooth trajectories. Overall, Bulldozer offers a practical, scalable enhancement to APF that improves reliability in cluttered environments and suggests promising directions for dynamic scenarios and further optimization integration.

Abstract

Path planning is a fundamental component in autonomous mobile robotics, enabling a robot to navigate from its current location to a desired goal while avoiding obstacles. Among the various techniques, Artificial Potential Field (APF) methods have gained popularity due to their simplicity, real-time responsiveness, and low computational requirements. However, a major limitation of conventional APF approaches is the local minima trap problem, where the robot becomes stuck in a position with no clear direction toward the goal. This paper proposes a novel path planning technique, termed the Bulldozer, which addresses the local minima issue while preserving the inherent advantages of APF. The Bulldozer technique introduces a backfilling mechanism that systematically identifies and eliminates local minima regions by increasing their potential values, analogous to a bulldozer filling potholes in a road. Additionally, a ramp-based enhancement is incorporated to assist the robot in escaping trap areas when starting within a local minimum. The proposed technique is experimentally validated using a physical mobile robot across various maps with increasing complexity. Comparative analyses are conducted against standard APF, adaptive APF, and well-established planning algorithms such as A*, PRM, and RRT. Results demonstrate that the Bulldozer technique effectively resolves the local minima problem while achieving superior execution speed and competitive path quality. Furthermore, a kinematic tracking controller is employed to assess the smoothness and traceability of the planned paths, confirming their suitability for real-world execution.

Figures (15)

• Figure 1: Conceptual illustration of the proposed bulldozer technique for eliminating local minima in artificial potential field–based navigation: (a) presence of local minima modeled as pothole-like traps, and (b) flattening of the potential landscape through the bulldozer operation.
• Figure 2: Flowchart of the proposed block-based local minima elimination strategy with a ramp mechanism. When a local-minimum block is detected, its potential is increased by a boost $\Delta V_k$; the first detected block receives the largest boost, and subsequent boosts decrease monotonically ($\Delta V_{k+1}<\Delta V_k$), creating a ramp-like potential rise that helps the robot escape minima traps.
• Figure 3: Illustration of potential field generation: (a) the environment map with circular obstacles and a target point, and (b) the resulting potential field where high values are centered around obstacles and low values surround the goal.
• Figure 4: Block-wise analysis of the potential field for the U-shaped obstacle map. The figure displays a grid-based segmentation where each block is labeled with its block number (in blue) and its corresponding average potential value (in red). The average values are computed over all pixels in each block. The goal point is within block number 85. Blocks with very high average potentials typically correspond to obstacle regions, whereas blocks with lower values indicate more favorable areas for navigation. This representation supports the heuristic method for local minima detection by enabling comparison of a block's potential with its neighbors. The U-shape creates an enclosed region where a local minimum trap is likely to form, highlighting the need for a robust identification strategy. It is clear that block number 55 is a local minima since it has lower values than its surrounding 8 blocks.
• Figure 5: A zoomed section showing the potential field distribution after applying the proposed Bulldozer technique with the ramp concept on the blocks shown in Figure \ref{['fig:u_obstacle_avg_blocks']}. Block 55, initially identified as a local minimum, is assigned a high potential value (1000), and surrounding blocks inside the U-shaped obstacle are gradually filled with decreasing values. This gradient forms a ramp-like structure that guides the robot out of the trap area while ensuring no new local minima are introduced. The robot can then easily navigate to the goal point located at block number 85.