Industrial Robots —
Types, Manufacturers and
How to Select the Right One

Robot selection is one of the decisions most commonly made on the basis of familiarity rather than suitability. The integrator uses the brand they know, the manufacturer specifies the robot that was in the last project, and the result is an articulated robot performing a task that a SCARA would execute at twice the speed for two-thirds of the cost — or a cobot installed where a full industrial robot would have been faster, safer and cheaper to integrate over its lifetime.

This article covers the principal robot architectures and their performance characteristics, the major manufacturers and their flagship model families, and a practical framework for matching robot type to application. It is aimed at engineers and project managers who need to make or inform robot selection decisions without a dedicated robotics background.

Robot Architectures — What Each Type Is Good For

Articulated Robots (6-Axis)

The dominant architecture in industrial robotics. Six rotary joints arranged in a kinematic chain give the robot six degrees of freedom — sufficient to position the end-effector at any point within the work envelope and in any orientation. This flexibility makes articulated robots the default choice for tasks that require arbitrary tool orientation: welding, painting, machine tending, assembly, and material handling.

Payload ranges from sub-kilogram (small assembly and dispensing) to over 2,000kg for heavy automotive handling. Reach typically 500mm to 4,000mm. The trade-off for flexibility is mechanical complexity, higher unit cost versus simpler architectures, and a slower maximum cycle time than parallel-link designs for simple pick-and-place operations.

SCARA Robots (Selective Compliance Assembly Robot Arm)

Four-axis architecture with two horizontal rotary joints and one vertical linear axis plus rotation. Inherently stiff in the vertical direction (resisting downward forces during assembly insertion) and compliant horizontally (allowing the tool to self-locate in a hole or feature). This makes SCARA the optimal choice for insertion tasks — PCB assembly, fastener driving, pharmaceutical blister packing, and precision component assembly.

Typical payload 1–20kg, reach 150–1,000mm, cycle times significantly faster than equivalent articulated robots on pure pick-and-place. Not suitable where arbitrary tool orientation is needed — the end-effector axis is always vertical.

Delta (Parallel-Link) Robots

Three (or four) arms connected in parallel to a single moving platform, driven from motors mounted on a fixed base. The parallel kinematic structure allows very high acceleration and deceleration — delta robots are the fastest architecture for lightweight pick-and-place, routinely achieving 150–200 picks per minute that no articulated robot can match at similar payload.

Payload 1–15kg typically, limited to a relatively shallow work dome rather than a full spherical envelope. Standard application is high-speed food, pharmaceutical and electronics packaging — sorting, grouping, and placing small items from a moving conveyor. Not appropriate where the load or reach requirements exceed the architecture's limits.

Cartesian / Gantry Robots

Three linear axes (X, Y, Z) assembled into an orthogonal structure. The mechanical simplicity produces high rigidity, high positional accuracy, and very high payload capacity for the cost. Gantry robots spanning large areas can handle tonnes. They are not flexible — the Cartesian structure cannot reach around or underneath a workpiece — but for applications that do not need that flexibility (CNC machine loading, laser cutting, large-format pick and place, storage and retrieval), Cartesian design is often the most cost-effective approach.

Often not considered "robots" in the traditional sense but they compete directly with 6-axis robots in many material handling applications and should be evaluated alongside them.

Collaborative Robots (Cobots)

Articulated robots (typically 6-axis) designed to operate near or alongside humans without a safety fence, relying on force-torque sensing, speed and position monitoring, and power/force limiting to detect contact and stop before injury occurs. ISO/TS 15066 defines the collaborative operation modes and the contact force limits that govern cobot design.

Key characteristics and limitations:

• Payload: Most cobots are rated 3–25kg. A small number reach 35kg. This covers the vast majority of manual assembly tasks but excludes heavy material handling.
• Speed: Cobots run at reduced speed compared to industrial robots — typically 1–2 m/s maximum TCP speed versus 3–5+ m/s for industrial. In practice, the maximum speed in collaborative mode (with humans present) is constrained further by the ISO/TS 15066 contact force limits. A cobot operating at full collaborative speed is substantially slower than an equivalent industrial robot in a fenced cell.
• Programming: Cobots are typically simpler to program than industrial robots — drag-to-teach, tablet interfaces, and graphical programming environments lower the barrier to deployment. This genuinely reduces integration cost for simpler applications.
• Risk assessment: Despite the marketing, cobots do not eliminate the need for safety risk assessment. ISO/TS 15066 requires a risk assessment to determine what speed and contact force limits are appropriate for the specific application, including any tooling or workpiece hazards. A cobot with a sharp tool or a heavy payload is not inherently safe to operate without guarding at full speed.

Key Specifications

Major Manufacturers and Key Model Families

FANUC (Japan) — Market Leader by Volume

FANUC is the world's largest robot manufacturer by installed units and dominates the automotive, electronics, and general industry sectors globally. Their control systems are regarded as class-leading for reliability, and the FANUC ecosystem — controllers, drives, vision systems, programming software — is deep and well-supported. Parts availability and service coverage in the UK is strong through FANUC UK.

Key model families:

• LR Mate 200iD — small 7kg articulated robot, 717mm reach. The benchmark small robot for machine tending, assembly and dispensing. Very fast cycle time for its class.
• M-10iA / M-20iD — mid-range articulated, 12–20kg payload. General purpose welding, assembly, handling. Huge installed base in UK manufacturing.
• M-710iC — 50–70kg articulated. Heavy machine tending, material handling, palletising at moderate rates.
• M-410iB/iC — 160–700kg dedicated palletising robot with low, wide work envelope optimised for layer palletising. Among the most widely deployed palletisers in food and FMCG.
• CR series (CR-4iA to CR-35iA) — FANUC's collaborative line, 4–35kg payload. The CR-35iA is one of the highest-payload cobots available. Uses the standard FANUC R-30iB controller.
• P-series — painting robots with hollow wrist for paint hose routing. Used by virtually every automotive OEM.
• Genkotsu / SR series — SCARA robots for electronics and small assembly.

ABB Robotics (Sweden) — Strongest in Arc Welding and Process

ABB has one of the widest robot portfolios across all industrial sectors, with particular strength in arc welding, automotive body-in-white, painting, and material handling. Their RobotStudio offline programming software is among the most capable and widely used in the industry. Strong UK service and support network.

• IRB 1200 — compact 5–7kg, 700–901mm reach. Small assembly, electronics, medical devices.
• IRB 2600 — 12–20kg, 1,650mm reach. The workhorse arc welding and machine tending robot. Excellent path accuracy for welding.
• IRB 4600 — 20–60kg. Assembly, press tending, palletising.
• IRB 6700 — 150–300kg. Automotive spot welding, heavy assembly, foundry.
• IRB 360 FlexPicker — 1–8kg delta robot. The reference product for high-speed pick and place in food, pharmaceutical and packaging. Up to 200 picks/min on the fastest variants.
• IRB 14000 YuMi — dual-arm 7-axis collaborative robot, 0.5kg per arm. Specifically designed for small parts electronics assembly alongside humans. Unique architecture — no other major manufacturer offers a comparable dual-arm cobot at this scale.
• GoFa CRB 15000 — 5kg cobot. ABB's main single-arm collaborative offering.

KUKA (Germany, now Midea Group) — Automotive Benchmark

KUKA is the robot of choice in European automotive body-in-white and the global benchmark for large, high-payload articulated robots. Their acquisition by Midea (China) in 2016 introduced uncertainty about long-term European manufacturing, but product development has continued. KUKA.Sim offline programming is well-regarded. UK support through KUKA Robotics UK.

• KR AGILUS — 6–10kg, 706–1,101mm reach. Fastest small industrial robot in the KUKA range. Machine tending, assembly, testing.
• KR CYBERTECH — 8–22kg. Arc welding, machine tending, assembly. The KUKA equivalent of the ABB IRB 2600 / FANUC M-10iA class.
• KR QUANTEC — 90–270kg. Automotive spot welding, press tending, heavy assembly. The robot in virtually every European car plant.
• KR FORTEC / TITAN — 300–1,300kg. Very large handling and press applications.
• LBR iiwa 7 / 14 — 7–14kg torque-controlled collaborative robot with force sensing in all joints. Used for precision assembly tasks requiring contact force control. More expensive to integrate than simpler cobots but offers genuine force-guided assembly capability.

Yaskawa Motoman (Japan) — Strong in Welding and Handling

Yaskawa's Motoman robots have a particularly strong position in arc welding (including their multi-axis welding systems with external axes) and in palletising. The Sigma servo drives and YASKAWA controller are reliable and well-regarded. Good UK support through Yaskawa Europe.

• GP series — general purpose articulated, 7–500kg. Wide range covering the full spectrum of industrial tasks.
• AR series — arc welding optimised. Hollow arm for cable management, torch cooler integration.
• PL series — dedicated palletising robots, low wide work envelope. Directly comparable to FANUC M-410.
• HC series (HC10DT, HC20DT) — collaborative robots 10–20kg. Skin-based contact detection (not force-torque sensing) — a different safety approach with implications for risk assessment.

Universal Robots (Denmark) — Cobot Market Leader

Universal Robots invented the modern cobot market and remains its largest player by volume. The UR ecosystem — UR+ marketplace of third-party end effectors and peripherals, large installed base, extensive integrator network — gives UR a significant practical advantage in cobot deployments. Programming via the teach pendant or PolyScope software is genuinely accessible compared to traditional industrial robot programming.

• UR3e — 3kg, 500mm reach. Bench-top assembly, dispensing, testing.
• UR5e — 5kg, 850mm reach. The most widely deployed cobot globally. Light assembly, machine tending, packaging.
• UR10e — 10kg, 1,300mm reach. Palletising at low rates, heavier assembly, welding.
• UR16e — 16kg, 900mm reach. Higher payload at moderate reach.
• UR20 — 20kg, 1,750mm reach. Palletising and large workpiece handling.
• UR30 — 30kg, 1,300mm reach. Heaviest standard UR. Machine tending for larger components.

Stäubli (Switzerland) — Precision and Cleanroom

Stäubli occupies a specific and valuable niche in applications where contamination, chemical resistance, or extreme precision is critical. Their sealed-arm robots (fully sealed, no external cabling) are the reference product for pharmaceutical manufacturing, semiconductor fabrication, and food processing requiring the highest hygiene standards. Not the right choice for general industry — the premium is significant — but the right choice where the application demands it.

• RX series — 6-axis, ISO Class 5 cleanroom rated. Electronics and pharmaceutical assembly.
• TX2 series — compact, high-precision 6-axis. Medical device, optics.
• TP80 Fast Picker — delta robot for high-speed food and pharmaceutical packaging.

Epson Robots (Japan) — SCARA Specialists

Epson's robot division produces some of the best SCARA robots available for electronics assembly and small parts handling. Their SCARA range is comprehensive from 3kg to 20kg, and their controller integration with vision systems is strong. A rational choice for any SCARA application in the sub-5kg payload range.

Application-Based Selection Guide

Collaborative vs. Industrial — The Real Decision

The cobot vs. industrial robot decision is frequently framed incorrectly. The question is not "do we want to work alongside the robot?" — it is "what is the lowest total cost system that meets the production requirement?"

Cobots are genuinely cost-effective when:

• The application is flexible and changes frequently — cobot reprogramming is genuinely faster than industrial robot reprogramming in most cases
• The production volume is low to medium — the cycle time penalty of cobots matters less when the robot is not the bottleneck
• Floor space is limited — cobots without fencing can be deployed in compact cells
• The integration complexity is low — UR+ ecosystem and similar marketplaces give cobots a real plug-and-play advantage for simple applications
• Payload is under 10kg — where the cobot's speed and payload are adequate, the lower unit cost and integration cost often wins

Industrial robots are the correct choice when:

• Cycle time is the constraint — an industrial robot in a fenced cell will outrun any cobot by a significant margin on most tasks above simple material transfer
• Payload exceeds 20–25kg — above this, industrial robots are the only viable option outside of a small number of specialist heavy cobots
• The application runs continuously and at high utilisation — industrial robots are designed for this; cobots at reduced collaborative speed running 24/7 may not recover their cycle time disadvantage
• Environmental conditions require it — high-temperature, corrosive, or ATEX-rated applications are handled by the industrial robot product lines; cobot variants for these environments are limited

Common Selection Mistakes

1. Ignoring tooling weight in the payload calculation. A robot rated at 10kg carrying a 6kg gripper has 4kg available for the workpiece. Robot selection with the gross payload figure and a notional tool weight produces undersized robots that either run below rated speed or fail prematurely.
2. Selecting on reach rather than work envelope shape. A robot may have the reach to access a target point but the geometry of the work envelope — the shape of the reachable volume — may mean it cannot reach it in a useful posture. Check the manufacturer's reach diagrams against the actual cell geometry in simulation before selecting.
3. Assuming cobots do not require risk assessment. CE marking under the Machinery Directive and ISO 10218-2 compliance requires a risk assessment for every robot installation, cobot or otherwise. The risk assessment determines what protective measures are needed — the cobot is the starting point, not the conclusion.
4. Not considering the controller ecosystem. The robot controller, not the mechanical arm, determines how the robot integrates with the wider cell — PLC communication, vision system interfaces, force-torque sensor integration, offline programming capability, remote diagnostics. Two robots with similar mechanical specifications may differ enormously in integration complexity depending on the controller.
5. Selecting a robot without considering spare parts and service availability. In the UK, FANUC, ABB, KUKA, Yaskawa, and Universal Robots all have established service and parts networks. Less well-known brands may offer attractive unit prices but carry real risk of extended downtime if the local support infrastructure is thin.
6. Specifying a 6-axis robot where a SCARA would be faster and cheaper. For pure vertical-pick-and-place assembly in a single horizontal plane, a SCARA will outperform a 6-axis robot on cycle time and usually cost less. The flexibility of 6-axis is wasted and the inertia of the additional axes costs speed.

Summary

Robot selection starts with application analysis, not brand preference. Define the payload (including tooling), the reach envelope, the required cycle time, the environmental conditions, the IP requirement, and the integration constraints before looking at products. Then select the architecture that is physically suited to the task — delta for high-speed lightweight pick-and-place, SCARA for vertical insertion assembly, articulated for everything that requires spatial flexibility — and within that architecture, select the manufacturer on the basis of application fit, controller ecosystem, and local support.

The cobot decision is a commercial and operational question, not a technical one. If the cycle time is achievable, the payload is within range, and the flexibility and integration cost savings justify the slower throughput, cobots are the right choice. If any of those conditions are not met, an industrial robot in a properly assessed cell is usually the correct answer.