Design and Implementation of a Snake - like Robot with Compact Joint Hinges
My study aims to design and implement a 14-joint, 2-DOF (degree-of-freedom) hinge-type snake-like robot. The robot adopts a modular design, with each active unit driven by dual-DOF servos, enabling serpentine motion, rolling, and crawling gaits. By controlling servo angles via sine-wave functions, combined with steering logic and mechanical optimization, we achieve a highly adaptable robot capable of operating in diverse environments, particularly for cable trench and pipeline inspection.
Features of My Snake-Like Robot
Research Problem and Hypothesis
How to reduce weight and size of 2-DOF modular hinge joints while maintaining/enhancing torque output, and integrate sensors/vision for broader applications?
Optimize single-joint internal structure with a two-stage gear reduction system to increase transmission ratio, direct servo torque to joints, reduce diameter/weight, and add vision sensors for navigation and image transmission.
Research Content
Two Gears C-shaped module design with compact hinges
Design 2-DOF modular spherical joint layout.
Modular joint internal structure design with gear reduction for torque enhancement.
Multi-joint serial connection with vision sensors
Stage 1:
Complete 14-joint prototype design, demand analysis, and 7 spherical joints (14 DOF).
Stage 2:
Optimize 2-DOF joints, improve torque, and test steering.
Stage 3:
Plan gaits (crawling, rolling, serpentine), implement algorithms/vision on embedded processors, and design V2 robot.
U-Shaped Joint Design with Double Gears
To meet the high torque output requirement, a gear reduction design was implemented. As shown, the two servo brackets are vertically aligned, connecting to servo discs oriented along the X-axis and Y-axis. Power is transmitted from the servo discs to the 12-tooth gear, then to the 26-tooth gear, and finally to the U-shaped connectors via flange bearings for output.
Gait Algorithm With Physics Principle
• The snake-like robot adopts a modular design with fourteen joints (seven motion units), each driven by a 2-DOF servo motor. The servo rotation range is 0-1000 (originally corresponding to 0-240°, but adjusted to 0-200° after simulation to prevent mechanical interference).
• To achieve serpentine, rolling, and crawling gaits, we employ sine waves (replacing traditional serpentine curves) to control the servo angle variations.
System Flowchart
Static Force Simulation for Single Module
Stress analysis revealed high von Mises stress concentrations at gear contact points and edges, indicating these areas bear significant load during power transmission. These high-stress zones may compromise gear durability and reliability, necessitating material upgrades to improve wear resistance and fatigue strength in subsequent optimizations.
Experiment and Data Collection/Analysis
Building on the robot's all-terrain capability and multi-modal motion features described in the mechanical design section, we conducted performance tests for all three gaits. To demonstrate motion efficiency, experiments were performed on three surface materials (indoor smooth floor, acrylic board, outdoor rough ground), with sand added as an extreme environment test case. These tests validated the robot's multi-environment mobility.
•Smooth Indoor Floor:Optimal for rolling gait (low friction enhances tumbling)Crawling efficiency decreases at high frequency
•Acrylic Board:Excellent serpentine gait performance at high frequencyReduced crawling efficiency at high frequency
•Wooden Board: Superior serpentine gait efficiency (especially high frequency)Stable rolling performance (moderate friction benefits motion
Conclusion and Future Plan
• The is study successfully designed and implemented a 7-joint, 2-DOF snake-like robot capable of serpentine, rolling, and crawling gaits. By controlling servo angles through sine-wave functions combined with steering control and mechanical optimization, I developed a highly adaptable robot for multi-environment operations.
• For future research, I plan to develop a third-generation robot focusing on obstacle navigation and climbing abilities to tackle more complex tasks. Specifically, the next version will achieve wall and tree climbing through innovations in mechanical design, motion control algorithms, and sensor integration.
Haiyuan Li’s robot that has the ability to climb stairs
Video for Gait Display