Planning Strategy for the Omni-directional Movement Path of the Four-wheel Differential Robot

Xue Wang, Wei Ma, Yiran Chen2012 · DOI

Abstract

The four-wheel differential omni-directional mobile robot (4WDR) has shown significant research value and application potential due to its high maneuverability. This paper explores the kinematics and dynamics of such robots, specifically focusing on a novel path-planning strategy. By utilizing dual reference points and deriving mathematical "temporary transition paths" for linear and circular trajectories, the proposed method ensures smooth and stable movement even when the robot starts far from its target. This research provides a theoretical basis for the advanced control of omni-directional mobile robots.

Introduction: The Challenge of 4WDR Control

Wheeled Mobile Robots (WMRs) are fundamental to modern automation, but four-wheel differential drive systems present unique control challenges. These robots typically consist of two independent differential units capable of 360° rotation around the robot body, providing extreme maneuverability.

Figure 1: Structural configuration of the four-wheel differential robot body, highlighting the differential units and casters.

However, standard path-following algorithms often struggle when the robot’s initial position is significantly offset from the desired path. In these scenarios, controllers may attempt to "snap" to the path too aggressively, leading to "excessive control," mechanical stress, or jerky movements. To address this, a geometric "on-ramp"—a temporary path—is required to allow the robot to merge into its intended trajectory smoothly and predictably.

Kinematic Modeling and Dual Reference Points

The proposed method simplifies the 4WDR by focusing on the two active differential units, treating the four casters as passive support elements. A world coordinate system and a joint coordinate system are established to track the position, posture, and heading angle (\θ) of these units.

A key innovation in this strategy is the use of dual reference points. Instead of using the robot's geometric center as a single reference, the authors designate the centers of the front and rear differential units as primary reference points. The control logic dictates that both points follow the same path, with the rear point following the front one with a calculated time delay. This approach effectively treats the robot like a two-car train, significantly simplifying the coordination logic required for complex maneuvers.

Temporary Path Generation

The core of the strategy lies in the mathematical derivation of temporary transition paths for different target geometries:

Linear Paths

The authors categorize potential straight-line targets into eight distinct cases based on their orientation and coordinate intercepts. By using a combination of counterclockwise arcs and line segments, the system creates a smooth transition from the starting point to the target line. This ensures that the robot enters the target path tangentially, avoiding abrupt changes in heading.

Figure 4: Geometric construction of temporary transition paths for linear target trajectories (Cases 1-4).

Circular Paths

For circular trajectories, the strategy accounts for whether the robot starts inside or outside the desired circle. The transition involves rotating the reference point to a tangent line parallel to the circle's center-origin vector before merging. This geometric approach prevents the robot from oscillating as it attempts to find the correct orbital radius.

Key Insights and Results

  • The "On-Ramp" Philosophy: Path planning is redefined to include not just the destination path, but the transition to it. This mitigates the risk of control saturation and instability.
  • Simplification through Symmetry: While eight categories for straight-line paths are identified, many are symmetrical. This allows complex navigation problems to be solved using a few core geometric templates that are mirrored or rotated.
  • Lag-Based Control: Forcing the rear differential unit to follow the front unit's path with a time delay reduces a multi-body coordination problem into a simpler "follow-the-leader" geometry.

The resulting mathematical framework provides a complete set of equations for calculating the position and posture of the robot during both standard movement and temporary transitions. The authors conclude that these paths are optimal and provide a necessary theoretical foundation for real-time controllers.

Conclusion and Future Work

This research establishes a robust geometric framework for 4WDR path planning. However, the current models are primarily theoretical, assuming "pure rolling" without slipping and treating all components as perfectly rigid bodies. Future work will involve implementing these strategies in simulation environments (such as ROS) and on physical hardware to validate performance under real-world dynamic constraints, including floor friction and sensor noise.

Summary updated

Abstract

The four-wheel differential omni-directional mobile robot has shown a strong research value and application prospect. This paper explored the kinematics and dynamics model of the omni-directional mobile robot by former differential and rear differential features, with a particular focus on its path planning strategy. The design includes path planning reference point, whilst analysis was made on the temporary path design strategy for linear and circular two features, with the resulting temporary path being optimal and most valid. This paper concludes with the research results that are available for providing the necessary preliminary studies and theoretical basis for the control of the mobile robot of this type.
BibTeX
@misc{wang2012planning,
  author = {Xue Wang and Wei Ma and Yiran Chen},
  title = {{Planning Strategy for the Omni-directional Movement Path of the Four-wheel Differential Robot}},
  year = {2012},
  doi = {10.2991/emeit.2012.295},
}
Figure