Robotics

Chapter 3
Robot Mechanics and Mechatronic Design

Bachelors Program in Machine Technology
Robotics   •   Bachelor

Robot Mechanics and Mechatronic Design

Introduction and Context

Robot mechanics concerns how a robot’s geometry, joints, links, actuators, transmissions, structure, sensors, and control system combine to produce useful motion. Mechatronic design extends purely mechanical design by treating mechanics, electronics, sensing, control, software, safety, and manufacture as one interdependent system.

A robot may be considered successful only when it can move accurately, safely, reliably, and economically under its intended load. A mechanically elegant arm with excessive backlash cannot achieve precise assembly. A powerful actuator mounted on a flexible link may produce poor accuracy. A rigid mechanism with an unsuitable controller may vibrate or become unstable. Robot design is therefore a systems problem rather than a sequence in which mechanical design is completed before electronics and software are added.

This chapter examines serial, parallel, and hybrid mechanisms; joint and link selection; transmissions; stiffness and backlash; payload and workspace; materials and machine elements; and a systematic mechatronic design process.

Learning Objectives

By the end of this chapter, a student should be able to:

  1. Distinguish serial, parallel, and hybrid robot architectures.
  2. Explain the function and limitations of common joints, links, actuators, and transmissions.
  3. Calculate basic torque, payload, speed, stiffness, and gear-ratio requirements.
  4. Analyse workspace, singularities, backlash, compliance, and error sources.
  5. Select materials and machine elements using load, stiffness, wear, manufacturability, and safety criteria.
  6. Apply a structured mechatronic design process to a robot application.

1. Robot Mechanisms as Kinematic Chains

A mechanism is an arrangement of rigid or compliant bodies connected so that relative motion is constrained. A robot is a mechanism whose motion is usually actuated and programmable. The bodies are called links, and the connections are joints or kinematic pairs.

A kinematic chain becomes a manipulator when one link is fixed as the base, selected joints are actuated, and an end-effector is attached. The joint variables describe configuration. For a revolute joint, the variable is an angle θ; for a prismatic joint, it is a displacement d.

Degrees of freedom

Degrees of freedom (DOF) describe the number of independent coordinates required to specify a mechanism’s configuration. In three-dimensional free space, a rigid body has six DOF: three translations and three rotations. A robot may have fewer than six task DOF, exactly six, or more than six (redundant motion).

For a planar mechanism, a commonly used mobility relation is:

M = 3(n − 1) − 2j₁ − j₂

Here M is mobility, n is the number of links including the fixed link, j₁ is the number of one-DOF joints such as revolute or prismatic joints, and j₂ is the number of two-DOF joints. The equation assumes planar motion and generic, independent constraints. Special geometries, redundant constraints, or coincident axes can invalidate the simple count.

A six-axis articulated robot commonly has six revolute joints, allowing independent control of three position coordinates and three orientation coordinates over part of its workspace. A four-axis SCARA robot is optimised for planar assembly and vertical insertion rather than arbitrary spatial orientation.

2. Serial, Parallel, and Hybrid Architectures

Serial robots

A serial robot consists of links connected end-to-end from base to wrist. Its forward kinematics are obtained by composing transformations from one joint to the next. Serial architecture offers a large and relatively unobstructed workspace and excellent flexibility. Industrial articulated arms, Cartesian robots, cylindrical robots, and SCARA robots are important examples.

Its disadvantages are cumulative: downstream links carry the mass of upstream links and payload; errors and deflections propagate to the end-effector; and the structure often becomes less stiff when fully extended. The actuator at the base may require a high torque because it supports all subsequent masses.

Parallel robots

A parallel robot connects the base to the platform through two or more independent kinematic chains. Delta pick-and-place robots and Stewart platforms are examples. Loads are shared among branches, producing high stiffness-to-mass ratio and rapid motion. However, the workspace is often smaller, singularities may be difficult to avoid, and calibration and inverse kinematics can be more demanding.

A Stewart platform has six variable legs and can control three translations and three rotations of a moving platform. Its legs usually experience axial tension or compression, but joint alignment and actuator accuracy are critical.

Hybrid robots

Hybrid robots combine serial and parallel submechanisms. A serial positioning stage may carry a parallel wrist, or a parallel mechanism may be mounted on a serial column. Hybrid designs attempt to combine broad positioning workspace with local stiffness, speed, or precision. They also inherit the modelling, calibration, and maintenance challenges of both architectures.

CriterionSerialParallelHybrid
WorkspaceUsually largeOften limited or shapedApplication-dependent
StiffnessModerate to low at extensionGenerally highCan be selectively high
Payload distributionMostly carried sequentiallyShared among branchesMixed
CalibrationRelatively straightforwardOften sensitiveMore complex
Typical useGeneral manipulationHigh-speed pick-and-place, simulationPrecision and specialised tasks

Architecture should be selected from task requirements, not from a general belief that one topology is superior.

3. Joints, Links, and Workspace

Joint types

A revolute joint permits rotation about an axis. It is compact and common in articulated arms. A prismatic joint permits translation along an axis and is used in Cartesian robots and telescoping axes. Other joints include cylindrical, spherical, universal, and screw joints. In practical robots, bearings, seals, shafts, couplings, and sensors are used to realise these idealised joints.

Joint selection must consider range of motion, speed, torque, radial and axial load, friction, sealing, stiffness, life, encoder mounting, and maintenance. A bearing is not automatically an adequate joint: bearing life, shaft deflection, housing stiffness, preload, and misalignment must also be evaluated.

Links

Links transmit forces and define geometry. Their design involves a compromise among mass, stiffness, strength, inertia, manufacturability, and cost. A link can be strong enough against static failure yet too flexible for the required accuracy. Hollow aluminium or carbon-fibre sections can provide high stiffness-to-mass ratio, but joints and interfaces often dominate the total compliance.

Workspace

The workspace is the set of positions and orientations reachable by the end-effector under joint limits, collision constraints, and actuator limits. The reachable workspace includes all attainable points; the dexterous workspace is the subset in which required orientations can also be achieved.

For a planar two-link arm with link lengths l₁ and l₂, the radial reach is bounded by:

|l₁ − l₂| ≤ r ≤ l₁ + l₂

where r is the distance from the base joint to the end point. This ideal result assumes full rotational freedom and no collision or joint-limit restrictions. Real workspace is smaller because of mechanical stops, self-collision, tooling, cables, and singularity avoidance.

A singularity is a configuration in which the robot loses instantaneous motion capability in one or more task directions, or requires very large joint velocities to produce a small Cartesian velocity. Near a singularity, force transmission, numerical conditioning, and control performance can deteriorate. Trajectories should therefore be planned with adequate distance from known singular configurations.

4. Actuators and Transmissions

Actuators convert energy into mechanical motion. Electric servo motors dominate many industrial robots because they offer high controllability, cleanliness, and compatibility with digital drives. Brushless permanent-magnet motors are common. Pneumatic actuators are attractive for simple, compliant, or hazardous environments, whereas hydraulic actuators provide high force density but require pumps, valves, seals, and fluid management.

A transmission changes speed, torque, direction, or the location of actuation. Common choices include:

  • harmonic or strain-wave drives: high reduction ratio and low backlash, but limited shock tolerance and torsional compliance;
  • planetary gearboxes: compact, efficient, and capable of high torque density;
  • cycloidal drives: robust against shock and suitable for high reduction, with manufacturing complexity;
  • belt drives: quiet and lightweight, but elastic and susceptible to tension loss;
  • ball screws: efficient rotary-to-linear conversion, but sensitive to contamination and alignment;
  • rack-and-pinion systems: suitable for long linear travel, with backlash and lubrication considerations.

For an ideal gearbox, the speed and torque relationships are approximately:

i = ωₘ / ωₗ
Tₗ = η i Tₘ

Here i is the reduction ratio, ωₘ and ωₗ are motor and load angular speeds, Tₘ and Tₗ are motor and load torques, and η is transmission efficiency. The equations assume steady operation and neglect gearbox inertia, friction variation, and transient shock. Units must be consistent, such as rad/s and N·m.

Worked example: motor selection

An axis requires 18 N·m at the output shaft and operates at 60 revolutions per minute. A gearbox has ratio i = 10 and efficiency η = 0.90. The required motor torque is:

Tₘ = Tₗ ÷ (ηi) = 18 ÷ (0.90 × 10) = 2.0 N·m

The motor speed is:

nₘ = inₗ = 10 × 60 = 600 rpm

The motor must provide at least 2.0 N·m continuously under the stated conditions. A practical selection also requires acceleration torque, peak load, thermal duty, reflected inertia, emergency-stop loads, and a safety margin.

5. Torque, Payload, and Dynamic Performance

For a force acting perpendicular to a lever arm, torque is:

T = F r

where T is torque in N·m, F is force in newtons, and r is perpendicular distance in metres. For a mass m lifted vertically, F = mg, where g ≈ 9.81 m/s² near Earth’s surface. If the force is not perpendicular, the general magnitude is T = Fr sin φ, where φ is the angle between the position vector and force vector.

A serial arm’s static joint torque is the sum of gravitational, payload, and other moments about that joint. A simplified horizontal reach estimate is:

T ≈ (mₚ + mₗ) g L

where mₚ is payload mass, mₗ is the equivalent downstream link mass, and L is the horizontal distance from the joint axis to the combined centre of mass. This is a simplified static equation; it excludes acceleration, friction, cable forces, and dynamic coupling.

Worked example: payload torque

A wrist carries a 4 kg tool and a 6 kg payload whose combined centre of mass is 0.35 m from the wrist axis. The static gravitational torque is:

T = (4 + 6) × 9.81 × 0.35 = 34.3 N·m

If a design factor of 1.5 is applied for specified uncertainty, the design torque is approximately 51.5 N·m. This factor is not a substitute for a dynamic analysis or manufacturer limits.

Dynamic motion adds inertial torque. For a rotating rigid body:

T = Jα

where J is mass moment of inertia in kg·m² and α is angular acceleration in rad/s². Total actuator torque may include gravitational, inertial, frictional, transmission, and disturbance components. Motor sizing must distinguish continuous torque from peak torque and account for thermal limits.

Payload is not simply the mass that a motor can lift. It depends on centre-of-mass location, orientation, acceleration, duty cycle, allowable deflection, joint life, and safety requirements. Manufacturers therefore specify payload together with reach, speed, and operating conditions.

6. Stiffness, Compliance, and Backlash

Stiffness is resistance to deformation. Linear stiffness is:

k = F ÷ δ

where k is stiffness in N/m, F is applied force in N, and δ is displacement in m. Rotational stiffness is similarly kθ = T ÷ θ, with units N·m/rad.

For components acting as springs in series, compliance adds:

1 ÷ keq = ∑(1 ÷ kᵢ)

where k_eq is equivalent stiffness and kᵢ are component stiffnesses. The softest component often dominates. For parallel load paths, stiffnesses add approximately, provided they undergo the same displacement and load sharing is valid.

Deflection causes endpoint error. A force of 100 N applied to a structure with stiffness 50,000 N/m produces:

δ = F ÷ k = 100 ÷ 50,000 = 0.002 m = 2 mm

Two millimetres may be unacceptable in precision assembly even though the structure has not yielded.

Backlash is lost motion caused by clearance between mating transmission elements. It creates direction-dependent positioning error and can cause oscillation when a controller reverses direction. Preloaded gears, roller screws, harmonic drives, direct-drive motors, and software compensation can reduce its effect. Compensation is limited when backlash varies with load, temperature, wear, or direction.

Accuracy is closeness to the commanded or true position. Repeatability is the ability to return to the same position. A robot may be highly repeatable but inaccurate because of calibration error. Stiffness and backlash affect both behaviour, but they should not be treated as identical phenomena.

7. Materials and Machine Elements

Material selection should begin with functional requirements rather than a preference for a particular material. Important criteria include density, yield strength, fatigue strength, elastic modulus, wear resistance, corrosion resistance, thermal expansion, damping, machinability, joining method, cost, and recyclability.

Steel offers high strength, stiffness, wear resistance, and economical manufacture, but its density increases inertia. Aluminium alloys reduce mass and are readily machined, but have lower elastic modulus and may require larger sections. Titanium provides high specific strength and corrosion resistance at high cost. Carbon-fibre composites offer excellent specific stiffness along selected directions, but require careful joint design and may have anisotropic failure modes. Engineering polymers are useful for covers, guides, gears, and vibration isolation, but temperature, creep, and wear must be assessed.

ElementMain design concerns
ShaftTorsion, bending, fatigue, keyway stress concentration, critical speed
BearingStatic and dynamic load, life, preload, contamination, misalignment
GearTooth bending, contact stress, lubrication, noise, backlash
BeltTension, fatigue, slip, pretension, pulley alignment
CouplingMisalignment, torsional stiffness, axial displacement, failsafe behaviour
FastenerPreload, joint separation, fatigue, corrosion, locking method
Linear guideMoment capacity, preload, contamination, rail alignment

For a solid circular shaft under torsion, the maximum shear stress is:

τmax = 16T ÷ (πd³)

where τ_max is shear stress, T is torque, and d is shaft diameter. This equation applies to a circular shaft under idealised elastic torsion; stress concentrations, keyways, combined bending, fatigue, and deflection may require more advanced analysis.

Machine elements should be designed as assemblies. A strong gear on a flexible shaft does not create a stiff transmission. A correctly selected bearing can fail prematurely if the housing is misaligned or contaminated. Fasteners require appropriate preload and locking, especially where vibration is present.

8. Systematic Mechatronic Design Process

A robust design process is iterative:

  1. Define the task. Specify payload, reach, cycle time, accuracy, repeatability, environment, duty cycle, safety category, and expected service life.
  2. Generate architectures. Compare serial, parallel, and hybrid alternatives using workspace, stiffness, cost, manufacturability, and maintenance criteria.
  3. Develop kinematic models. Establish coordinate frames, joint limits, forward and inverse kinematics, singularities, and collision constraints.
  4. Estimate loads. Calculate static and dynamic torque, forces, accelerations, inertia, and worst-case disturbance loads.
  5. Select actuators and transmissions. Match continuous and peak torque, speed, efficiency, thermal duty, resolution, backlash, and safety braking requirements.
  6. Design the structure. Check strength, fatigue, stiffness, natural frequencies, thermal distortion, and interfaces.
  7. Integrate sensing and control. Select encoders, limit switches, torque sensors, motor drives, communication networks, and control bandwidth.
  8. Prototype and verify. Test workspace, payload, repeatability, thermal behaviour, vibration, emergency stopping, and fault response.
  9. Iterate for production. Consider tolerances, calibration, assembly time, supply chain, service access, documentation, and end-of-life disposal.

Computer-aided design, multibody simulation, finite-element analysis, and digital control simulation are valuable, but their results depend on assumptions. A finite-element model with incorrect boundary conditions can provide precise-looking but misleading results. Physical testing remains essential.

9. Illustrative Case: High-Speed Pick-and-Place Robot

Background

A packaging line must transfer 1 kg containers from a conveyor to trays. The required horizontal workspace is 600 mm × 400 mm, the cycle time is 0.8 s, and the robot must maintain repeatability within ±0.5 mm. The environment is clean but includes frequent acceleration and deceleration.

Problem

A six-axis serial robot can reach the area, but its wrist and downstream links create high moving inertia. At high speed, vibration and settling time reduce throughput. A Cartesian robot offers simple control but would require heavy moving axes and a large frame.

Analysis

A delta-style parallel robot is considered because its actuators can remain near the base while lightweight arms move the end-effector. Its advantages are low moving mass, high acceleration, and good repeatability. The design team must nevertheless check workspace boundaries, leg interference, singularities, bearing life, belt or gearbox backlash, and the effect of payload-induced deflection.

The payload force under static conditions is approximately 9.81 N. If the end-effector accelerates at 5g vertically, the inertial force associated with the payload is approximately 49.1 N, excluding gravity and tool mass. This illustrates why dynamic design loads can greatly exceed static payload weight.

Approach and solution

The team selects a parallel mechanism with three actuated arms, brushless servo motors, preloaded joints, lightweight composite links, and an aluminium base. Encoders are mounted close to the actuators, while calibration fixtures establish geometric parameters. A compliant gripper is used to reduce impact during contact. The controller limits acceleration near workspace boundaries and avoids identified singular regions.

Outcome

Prototype testing confirms the required workspace and repeatability at nominal payload. At maximum acceleration, however, the end-effector vibrates. Increasing link stiffness, improving joint preload, and adding a short settling interval produces acceptable performance, although the cycle time is slightly increased. The final design meets the production requirement because mechanical stiffness and control tuning were treated together.

Learning points

The case demonstrates that architecture, mass distribution, transmission compliance, calibration, and control cannot be evaluated independently. The fastest nominal mechanism is not necessarily the most productive if it requires long settling time or frequent maintenance.

10. Safety, Standards, and Contemporary Developments

Robot design must address safeguarding, risk assessment, emergency stopping, unexpected restart, collaborative operation, pinch points, stored energy, and safe maintenance. Applicable national regulations and standards should be identified early; commonly referenced frameworks include industrial robot safety standards such as ISO 10218 and collaborative robot guidance such as ISO/TS 15066, subject to the jurisdiction and current edition.

Safety-rated monitored stop, speed and separation monitoring, force limitation, interlocked guarding, and redundant sensing may be required. A software limit alone should not be treated as the only protection against a hazardous motion. Risk assessment should consider foreseeable misuse, maintenance, calibration, cybersecurity, and loss of communication.

Contemporary development includes lightweight composite structures, direct-drive actuators, torque-controlled robots, additive-manufactured topology-optimised links, integrated motor-drive modules, and digital twins. Artificial intelligence can assist trajectory optimisation, anomaly detection, and predictive maintenance, but it does not remove the need for deterministic safety functions, validated limits, traceable data, and human supervision. AI-generated motion commands must remain within independently enforced mechanical and safety constraints.

Challenges and Limitations

Robot mechanics is limited by modelling uncertainty, manufacturing tolerances, temperature variation, wear, lubrication changes, cable drag, sensor errors, and unmodelled contact forces. Static equations may underestimate dynamic loads. Catalogue ratings may apply only under specified lubrication, speed, orientation, and life conditions. High stiffness can increase transmitted shock and reduce compliance during contact. Low mass can reduce inertia but may reduce damping or robustness. Parallel mechanisms reduce moving mass but can create complicated singularity and calibration problems.

Good engineering therefore combines analytical calculations, simulation, experimental identification, tolerance analysis, safety assessment, and lifecycle evaluation. Design margins should be justified rather than selected arbitrarily.

Chapter Summary

Robot mechanics links geometry, motion, force transmission, structural behaviour, and machine elements. Serial robots provide broad flexibility but accumulate mass, error, and compliance along the chain. Parallel robots provide high stiffness-to-mass ratio and rapid motion but generally have more complex workspaces and calibration. Hybrid mechanisms combine characteristics of both.

Joint and link selection must account for range, load, speed, stiffness, wear, and integration. Transmission ratio affects speed and torque, while efficiency and backlash influence practical performance. Payload calculations must include gravity, acceleration, centre-of-mass location, and duty cycle. Stiffness determines deflection, and compliance may dominate endpoint accuracy even when strength is adequate. Materials and machine elements should be selected through a systems analysis of strength, stiffness, mass, fatigue, environment, manufacture, cost, and safety.

A systematic mechatronic process begins with task requirements, compares architectures, models kinematics and loads, integrates actuation and sensing, verifies performance experimentally, and iterates toward manufacturable and safe production equipment.

Key Terms

Mechanism: A constrained arrangement of bodies that produces controlled relative motion.

Link: A member of a mechanism that transmits motion and force.

Joint: A connection that permits specified relative motion between links.

Degree of freedom: An independent coordinate required to define configuration.

Serial robot: A robot whose links form an end-to-end kinematic chain.

Parallel robot: A robot in which multiple chains connect the base to a moving platform.

Hybrid robot: A robot combining serial and parallel mechanisms.

Workspace: The set of positions and orientations reachable by the end-effector.

Singularity: A configuration in which instantaneous motion or force capability becomes degenerate.

Transmission: A machine arrangement that transfers and modifies motion and torque.

Backlash: Lost motion caused by clearance in a transmission.

Stiffness: Resistance to deformation under applied load.

Compliance: Displacement or rotation produced per unit load.

Payload: The useful mass carried by a robot under specified operating conditions.

Repeatability: Ability to return consistently to a location.

Accuracy: Closeness of achieved position to the intended or true position.

Inertia: Resistance of a mass distribution to acceleration.

Mechatronic design: Integrated design of mechanical, electronic, sensing, control, and software elements.

Singular configuration: A robot pose where the Jacobian loses rank or becomes poorly conditioned.

Duty cycle: The time-dependent pattern of operation, loading, and rest.

Review Questions

Short Answer Questions

  1. What is the principal structural difference between a serial and a parallel robot?
  2. Distinguish accuracy from repeatability.
  3. Define backlash and state one method of reducing it.
  4. What factors, besides payload mass, influence required actuator torque?
  5. Why can a robot be strong enough but insufficiently accurate?

Long Answer Questions

  1. Compare serial, parallel, and hybrid robot architectures in terms of workspace, stiffness, payload, calibration, and typical applications.
  2. Explain how joints, links, bearings, shafts, and transmissions combine to form a practical robot axis.
  3. Discuss a systematic procedure for selecting materials and machine elements for a robot arm.
  4. Explain the relationship among stiffness, compliance, backlash, accuracy, and repeatability.
  5. Describe how safety and control considerations should influence mechanical robot design from the beginning.

Analytical / Application Questions

  1. An axis must deliver 25 N·m at 100 rpm through a gearbox of ratio 8 and efficiency 0.85. Determine the ideal motor speed and the required motor torque, then identify two additional factors that must be checked before selecting the motor.
  2. A 5 kg payload is held 0.4 m from a joint axis. Calculate the static gravitational torque and explain why the actual design torque may be higher.
  3. A robot structure has two compliant components in series with stiffnesses 80,000 N/m and 20,000 N/m. Calculate the equivalent stiffness and the deflection under a 100 N load. Interpret the result for precision assembly.

Multiple Choice Questions with Answers

Question 1

Which characteristic is most generally associated with a serial robot?

A. Large and relatively unobstructed workspace

B. All loads shared equally by independent branches

C. No accumulation of positioning errors

D. No need for calibration

Correct Answer: A. Large and relatively unobstructed workspace

Question 2

For a planar two-link arm with full rotational freedom, the maximum radial reach is:

A. |l₁ − l₂|

B. l₁ + l₂

C. l₁l₂

D. l₁ ÷ l₂

Correct Answer: B. l₁ + l₂

Question 3

A gearbox has ratio i = 10, efficiency η = 0.9, and must deliver 18 N·m. What motor torque is required ideally under the stated model?

A. 0.2 N·m

B. 1.62 N·m

C. 2.0 N·m

D. 20 N·m

Correct Answer: C. 2.0 N·m

Question 4

Backlash is best described as:

A. Elastic deformation under load

B. Lost motion caused by clearance in mating transmission elements

C. Permanent material yielding

D. Motor heating during continuous operation

Correct Answer: B. Lost motion caused by clearance in mating transmission elements

Question 5

A 100 N force acts on a structure with stiffness 50,000 N/m. What is its deflection?

A. 0.0002 m

B. 0.002 m

C. 0.02 m

D. 2 m

Correct Answer: B. 0.002 m

Question 6

Which architecture commonly offers a high stiffness-to-mass ratio and rapid motion?

A. Parallel

B. Purely serial

C. Unactuated linkage

D. Flexible cable mechanism only

Correct Answer: A. Parallel

Question 7

The static torque from a 10 kg combined mass at a 0.35 m horizontal lever arm is approximately:

A. 3.43 N·m

B. 9.81 N·m

C. 34.3 N·m

D. 98.1 N·m

Correct Answer: C. 34.3 N·m

Question 8

Which statement correctly distinguishes repeatability from accuracy?

A. Repeatability concerns consistency; accuracy concerns closeness to the true position

B. They are always identical

C. Accuracy concerns only speed

D. Repeatability concerns only payload

Correct Answer: A. Repeatability concerns consistency; accuracy concerns closeness to the true position

Question 9

For springs or compliant elements in series, equivalent compliance is:

A. The sum of the individual compliances

B. Always zero

C. The product of all stiffnesses

D. Equal to the largest stiffness only

Correct Answer: A. The sum of the individual compliances

Question 10

Why should safety-rated functions not rely only on software position limits?

A. Software limits increase payload automatically

B. A software fault could leave hazardous motion insufficiently protected

C. Position limits prevent all mechanical wear

D. Software limits eliminate the need for risk assessment

Correct Answer: B. A software fault could leave hazardous motion insufficiently protected

Robotics
Chapter 3 of 15