Chronicle
Motion and Dynamics

Friction, Traction, and Stability

Understanding friction types, their effects on robot motion, and optimizing traction for different terrains

Friction, Traction, and Stability

Friction plays a paradoxical role in robotics: it's essential for movement and control, yet it also causes energy loss and wear. Understanding and managing friction is crucial for designing efficient, reliable robots.

Types of Friction


Friction Coefficients: Reference Table

Common Material Pairs:

Surface PairStatic (μ_s)Kinetic (μ_k)Rolling (C_r)Best Use
Rubber on Concrete0.9-1.00.7-0.80.01-0.02Standard wheel
Rubber on Ice0.15-0.250.1-0.150.002-0.005Poor traction
Metal on Metal0.4-0.60.3-0.4-Joint bearings
Teflon on Steel0.04-0.10.04-0.08-Low-friction joints
Wood on Wood0.4-0.50.3-0.4-Historical mechanisms
Rubber on Sand0.6-0.80.5-0.70.05-0.1Off-road wheels
Cleats on Grass1.2-1.50.8-1.00.03-0.05Terrain grip

Finding Coefficients

For materials not listed, you can:

  1. Search scientific literature
  2. Request from material suppliers
  3. Measure experimentally using an inclined plane
  4. Use approximate typical values (0.3-0.7 for most materials)

Traction and Mobility

Traction is the friction force that enables movement without slipping - essential for robot locomotion.

Maximum Traction Force

The maximum force a wheel can exert without slipping:

F_traction_max = μ × N = μ × m × g

Where:

  • μ = Coefficient of friction (typically μ_s for starting)
  • m = Robot mass
  • g = Gravity (9.81 m/s²)

Climbing Inclines

For a robot on a slope:

Example: 20 kg Robot on 30° Incline

Given: μ_s = 0.8 (rubber on concrete)

Step 1: Component parallel to slope
F_parallel = m × g × sin(30°) = 20 × 9.81 × 0.5 = 98.1 N

Step 2: Normal force perpendicular to slope
N = m × g × cos(30°) = 20 × 9.81 × 0.866 = 169.5 N

Step 3: Maximum available traction
F_max_traction = 0.8 × 169.5 = 135.6 N

Step 4: Can it climb?
F_max_traction (135.6 N) > F_parallel (98.1 N) ✓ YES, robot can climb!

Margin: 135.6 - 98.1 = 37.5 N (allows acceleration too)

Maximum Climbable Incline:

At the steepest angle, friction just equals the component pulling down:

μ × m × g × cos(θ) = m × g × sin(θ)
μ × cos(θ) = sin(θ)
tan(θ) = μ
θ_max = arctan(μ)

For μ = 0.8:

θ_max = arctan(0.8) = 38.7°

A robot with μ = 0.8 traction can climb up to 38.7° incline at zero velocity.


Traction Optimization Strategies

Increase Normal Force

Method: Increase robot weight/load on drive wheels

Ways to do this:

  • Add ballast/weight to drive wheels
  • Use weight distribution to bias wheels
  • Increase overall robot mass (for off-road)
  • Lower center of gravity

Trade-off: Heavier robots need more power to move, reducing efficiency and battery life

Best practice: Use minimum weight needed, concentrate it on drive wheels

Increase Friction Coefficient

Method: Use higher friction materials

High-friction wheel materials:

  • Rubber compounds (most common)
  • Silicone (very high friction)
  • Specialized track pads (extreme grip)

Surface improvements:

  • Tread patterns (increase micro-grip)
  • Textured surfaces (rubber dots)
  • Compliant wheels (deform to fit surface)

Trade-off: Softer, grippier materials wear faster

Best practice: Match material to expected terrain

Increase Contact Area

Method: Use larger or wider wheels

Effects:

  • Wider wheels: Better weight distribution, reduced sinking
  • Larger wheels: Better obstacle clearance, smoother motion
  • Tracks: Maximum contact area, best traction on soft ground

Trade-off: Larger wheels are heavier and slower

Best practice: Balance size with speed requirements and weight budget

Active Suspension

Method: Adjust weight distribution dynamically during motion

Techniques:

  • Shift internal weight (CG redistribution)
  • Active spring systems (adjust stiffness)
  • Lean compensation (in turning)

Advanced applications:

  • Terrain-adaptive suspension
  • Dynamic load balancing
  • Self-leveling platforms

Trade-off: Complex, heavy, power-consuming

Best practice: Use for demanding applications (rock crawlers, off-road competition)


Friction Management in Robots

Minimizing Unwanted Friction

In Joints and Bearings:

Specific Strategies:

ProblemSolutionEffect
Sliding surfacesBall/roller bearings50-70% loss reduction
Dry surfacesLubrication (oil/grease)30-60% loss reduction
Worn surfacesSurface treatment/coatings20-40% loss reduction
ContaminationSealed bearingsMaintains low friction

Energy Savings Example:

Robot arm with 50W friction loss:

  • Reduce by 50%: 25W savings (4× longer battery life!)
  • All-day operation vs. 1-hour operation

Lubrication Best Practices

  • Use recommended lubricant for each component
  • Over-lubrication can increase drag (thick films)
  • Under-lubrication causes rapid wear
  • Sealed bearings: Don't need relubrication
  • Open bearings: Relube every 50-100 operating hours

Maximizing Useful Friction for Gripping

Gripper Friction Requirements:

Required Friction Force = (Object Mass × g) / Number of Gripper Fingers

To prevent slipping of a 2 kg object with 2-finger gripper:

F_required = (2 × 9.81) / 2 = 9.81 N per finger

Strategies for Better Gripping:

  1. High-friction materials:

    • Rubber pads (μ ≈ 0.5-1.0)
    • Silicone (μ ≈ 0.8-1.5)
    • Specialized compounds (μ > 2.0)
  2. Textured surfaces:

    • Bumpy patterns increase micro-contact
    • Can increase friction 20-50%
  3. Compliant pads:

    • Soft materials conform to object shape
    • Increase contact area
    • Better for fragile objects
  4. Appropriate contact force:

    • Too little: Object slips
    • Too much: Damage to object or motor stall
    • Typical: 1.5-2× minimum required force

Friction Compensation in Control

Modern robots use sophisticated algorithms to handle friction:

Friction Modeling

Observed friction in robotics often follows:

τ_friction = τ_coulomb + τ_viscous × ω + τ_static × sign(ω)

Where:

  • τ_coulomb = Constant friction (Coulomb friction)
  • τ_viscous × ω = Velocity-proportional friction (viscous drag)
  • τ_static = Additional static friction at zero velocity

Feedforward Compensation

Robot controllers measure these friction components and apply compensating torque:

Benefits:

  • Smoother motion
  • No jerky stick-slip behavior
  • Better precision
  • Reduced energy waste

Trade-off: Requires accurate friction model (which varies with temperature, wear, contamination)


Summary: Friction in Practice

Key Takeaways:

✓ Rolling friction much less than sliding friction (use wheels!) ✓ Friction is essential for traction but wastes energy ✓ Lubrication reduces bearing friction 30-60% ✓ Higher friction materials improve gripping reliability ✓ Friction models enable precise motion control ✓ Maximum climbable incline = arctan(friction coefficient)

How is this guide?