Doosan H2017 Review: Specs, Payload & Reach

Doosan H2017: Quick Buyer Verdict

The H2017 sits in an interesting position in the collaborative-robot market. It combines the human-scale working envelope of a long industrial arm with a 20 kg payload and the force-sensing architecture expected from a modern cobot.

Its defining specification is not simply the payload. It is the combination of 20 kg capacity and 1,700 mm reach. That makes it relevant for applications where the robot must reach across machinery, service multiple positions, handle relatively heavy components or cover a larger palletising envelope without immediately moving to a conventional industrial robot.

Doosan also installs a high-performance torque sensor at each of the robot’s six joints. This enables collision detection, force-related safety functions, hand-guiding and force/compliance-control capabilities that are important in flexible automation.

However, the H2017 should not be bought from the specification table alone. Its current published repeatability is ±0.1 mm, the arm is IP54 rather than IP65 or IP67, installation is floor-only, and its approximately 1 m/s published TCP speed is conservative compared with some newer high-performance cobots.

Pros

• 20 kg payload.
• Long 1,700 mm working radius.
• Six-axis articulated design.
• High-performance torque sensor on every joint.
• Useful force and compliance-control capabilities.
• Configurable collaborative zones and safety limits.
• Strong fit for machine tending, palletizing and heavy material handling.
• Dart-Suite provides a comparatively approachable programming environment.
• Official ROS 2 ecosystem for advanced integration.
• 24 V tool power and integrated tool I/O.
• Controller options suitable for conventional cells and mobile-robot integration.
• One-year manufacturer warranty according to current Doosan service documentation.

Cons

• ±0.1 mm repeatability is not class-leading.
• IP54 protection is lower than IP65/IP67 competitors.
• Floor-only mounting restricts cell design.
• The arm is heavy at approximately 79 kg in the current V3 documentation.
• Published TCP speed is modest relative to newer high-performance cobots.
• 20 kg maximum payload can be restrictive once a heavy gripper, tool and cabling are included.
• Doosan does not publish a simple global online purchase price.
• A complete production cell still requires tooling, safety engineering, integration and commissioning.
• Newer Doosan and competitor models should be compared before committing to H2017.

Our recommendation: shortlist the H2017 if your application genuinely needs close to 1.7 m of reach and around 20 kg of payload while retaining torque sensing and collaborative-robot functionality. If the main requirement is pure palletizing, payload above 20 kg, high environmental protection or maximum cycle speed, compare alternatives before specifying the cell.

View the Doosan H2017 product page at Anton Robots for the current listing and enquiry options.

How Much Does the Doosan H2017 Cost in 2026?

Doosan Robotics does not publish a universal retail price for the H2017 on its current product page. The robot is sold through an enquiry and dealer/integrator process, so buyers should treat it as a quote-based industrial system rather than an off-the-shelf product with one global price.

This is important because the arm is only one part of the installed cost.

A working H2017 application may also require:

Do not compare arm prices

One of the easiest mistakes in cobot purchasing is comparing the price of one bare robot with another manufacturer’s complete application package.

For example, a palletizing quote may include the robot, gripper, safety scanner, base, software, commissioning and pallet-pattern interface. A cheaper quote may include only the arm, controller and pendant.

The correct comparison is:

Total cost to achieve the required production task at the agreed cycle time, payload, availability and safety level.

For broader market context, see our cobot price guide.

Request two prices

Ask the supplier for:

1. Robot package price: arm, controller, pendant, cables, software licences and standard accessories.
2. Turnkey cell price: everything required to reach production acceptance.

This makes comparison between Doosan, Universal Robots, FANUC and other suppliers considerably more meaningful.

What Is the Doosan H2017?

The Doosan H2017 is a six-axis collaborative robot designed for heavier industrial automation.

Its model name broadly reflects its two defining numbers:

• 20 kg payload.
• 1,700 mm maximum working radius.

The H2017 belongs to Doosan Robotics’ H-Series, which focuses on high-payload collaborative automation. The other current H-Series model, the H2515, trades some reach for greater lifting capacity: 25 kg payload and 1,500 mm reach.

The H2017 is therefore the long-reach member of the pair.

What the H2017 is

• A production-oriented six-axis collaborative robot.
• A long-reach cobot for industrial handling.
• A 20 kg payload robot for heavier components and larger tooling.
• A torque-sensing platform capable of force-aware operation.
• A robot that can be programmed through Doosan’s software ecosystem.
• A candidate for palletizing, machine tending, welding and material handling.
• An established commercial product with worldwide distributor support.

What the H2017 is not

• It is not automatically safe to run at full speed beside people.
• It is not a complete production cell by itself.
• It is not a 20 kg payload after ignoring the mass of the gripper and tooling.
• It is not an IP67 washdown robot.
• It is not designed for ceiling mounting according to the current H-Series manual.
• It is not the fastest cobot in the current heavy-payload market.
• It is not Doosan’s highest-payload palletizing robot in 2026.

If you are still choosing the robot category rather than one model, browse current collaborative robots before specifying a manufacturer.

Doosan H2017 Specifications

The following table reflects Doosan’s current H2017 product information and V3 robot documentation checked in September 2026.

A note about H2017 weight specifications

Older Doosan documentation can be found online showing H2017 arm weights around 72–74 kg. Current V3 documentation checked for this review lists approximately 79 kg.

For a new purchase, use the specification associated with the exact robot and controller revision being quoted rather than copying a number from an older datasheet.

For a used H2017, request the model revision, serial number and original manual before designing lifting equipment, a pedestal or a mobile platform around a weight figure.

Payload and Reach: Why the H2017 Exists

The H2017’s strongest argument is the combination of a relatively high payload and a long working radius.

A smaller cobot may have enough nominal payload for a part but fail to reach:

• the rear of a CNC machine;
• multiple machines from one base;
• both sides of a workstation;
• the far edge of a pallet;
• several conveyor positions;
• a large welding fixture.

A long-reach robot can reduce the need for linear axes, additional robots or complex repositioning.

20 kg does not mean a 20 kg workpiece

Payload calculations must include everything carried beyond the robot flange.

For example:

That example is already close to the H2017’s nominal limit before considering centre of gravity, acceleration, wrist moment and inertia.

Centre of gravity matters

A 15 kg object held close to the flange can create a very different load from the same 15 kg object held far forward by a long vacuum frame.

When requesting a feasibility check, provide:

• workpiece mass;
• tool mass;
• combined centre of gravity;
• tool geometry;
• required orientation;
• maximum acceleration;
• cycle time.

Do not approve an application from payload alone.

H2017 vs H2515

The closest internal alternative is Doosan’s H2515.

Choose H2017 when 200 mm of additional reach matters more than the extra 5 kg of H2515 capacity.

Choose H2515 when the complete tool and workpiece exceed the comfortable payload margin of the H2017.

Speed, Repeatability and Cycle Time

The H2017 is powerful, but it should not automatically be assumed to be a high-speed robot.

Doosan’s current model-specific documentation specifies approximately 1 m/s TCP speed, while individual joint speeds range from 80°/s to 180°/s.

That can be entirely adequate for machine tending, welding, heavy handling or collaborative operation. It may be less attractive for high-throughput transfer applications where raw cycle time dominates the purchasing decision.

Repeatability is ±0.1 mm

The published H2017 position repeatability is ±0.1 mm according to ISO 9283.

That is usually sufficient for:

• palletizing;
• case handling;
• machine loading;
• general material transfer;
• welding with appropriate process setup;
• many assembly operations with compliant tooling;
• camera positioning.

It may be less suitable when the process requires extremely tight robotic positioning without fixtures, vision correction or compliance.

Some competing cobots publish repeatability around ±0.04 or ±0.05 mm.

Robot speed is not production speed

A fast robot does not necessarily create a fast application.

Cycle time can be limited by:

• CNC machining time;
• gripper closing time;
• vacuum generation;
• part detection;
• conveyor timing;
• welding process speed;
• safety scanner zones;
• human access;
• PLC handshaking;
• payload and centre of gravity;
• required settling time.

For serious applications, request a simulation or witnessed cycle using the proposed tool and representative payload.

Doosan H2017 Safety and Collaborative Operation

The H2017’s six joint torque sensors are central to its collaborative architecture.

They allow the controller to monitor forces and torques throughout the arm rather than relying solely on motor-current estimation.

Doosan’s current safety documentation includes monitoring and control functions around:

• joint position;
• joint speed;
• joint torque;
• collision detection;
• TCP position;
• TCP orientation;
• TCP speed;
• TCP force;
• robot momentum;
• robot power;
• safety zones;
• reduced-speed operation.

Safety certification

Doosan’s product catalog lists certifications including KCs, CE, NRTL, EN ISO 10218-1 and EN ISO 13849-1.

Its documentation distinguishes between safety functions. Core functions such as emergency stop are documented at the higher PLe / Category 4 level, while many safety monitoring functions are rated PLd / Category 3 / SIL 2.

The important buyer takeaway is not the headline safety rating. It is that the integrator must choose and validate the appropriate functions for the application.

Can the H2017 operate without a fence?

Potentially—but not because it is called a collaborative robot.

A completed H2017 application can still require fencing or other separation when it involves:

• sharp tools;
• welding;
• heavy or unstable workpieces;
• high speeds;
• crushing points;
• dangerous machinery;
• large grippers;
• high-temperature parts;
• unsafe stopping distances.

The robot arm is only one component of the risk.

Collaborative Zones

Doosan allows zones to modify robot behaviour.

A collaborative zone can be configured to influence parameters such as speed, collision sensitivity, force limits and stop behaviour.

Other zone types can help manage:

• crushing hazards;
• tool orientation;
• restricted workspaces;
• areas requiring different collision sensitivity.

This is valuable in applications where the robot operates productively in one part of the cell but must become more conservative when it enters a human-accessible area.

Safety is part of the cycle-time calculation

An application that looks fast in an isolated robot demonstration may slow significantly once realistic safety separation and collaborative limits are applied.

Ask the integrator to provide:

• normal production cycle time;
• cycle time with a person inside the monitored area;
• protective-stop behaviour;
• restart procedure;
• stopping-distance calculations;
• validated safety configuration.

Programming, Dart-Suite and ROS 2

The H2017 can serve both conventional industrial users and more advanced software teams.

Dart-Suite

Doosan’s Dart-Suite ecosystem was created to make collaborative-robot programming more accessible.

The platform includes services around:

• application development;
• robot operation;
• application distribution;
• developer resources;
• community support.

Doosan describes Dart-IDE as a web-based development environment where robot functionality can be assembled using reusable components and graphical workflows rather than requiring every application to be built from low-level code.

For a factory buyer, the practical benefit is easier process changeover and less dependence on a specialist programmer for every minor adjustment.

Teach pendant

The H-Series supports Doosan’s teach pendant, which provides the normal industrial workflow for:

• jogging;
• teaching positions;
• building tasks;
• configuring robot parameters;
• setting safety limits;
• monitoring operation.

Doosan also offers its Cockpit direct-teaching control as an option for simplified position teaching.

ROS 2

Doosan maintains an official ROS 2 repository for its robot lineup.

The current package supports functions including:

• real robot control;
• virtual operation;
• RViz;
• Gazebo simulation;
• MoveIt 2;
• ros2_control;
• robot descriptions;
• example applications.

This makes the H2017 relevant not only to standard factory automation but also to laboratories, research groups and advanced integrators building custom planning or perception systems.

ROS 1 is now legacy

Doosan’s original ROS repository states that ROS 1 support ended in May 2025 alongside the broader end of ROS 1 support.

For a new 2026 project, use the current ROS 2 stack unless a legacy installation creates a specific reason not to.

Advanced integration is not the same as supported turnkey functionality

The existence of ROS 2, an API or a simulator does not mean a complex autonomous task is ready out of the box.

A custom application may still require:

• robotics software engineering;
• PLC integration;
• motion planning;
• vision;
• calibration;
• collision modelling;
• recovery logic;
• production testing.

Controllers, Tooling and Integration

A cobot becomes useful only after it is connected to the rest of the process.

Controller options

Doosan lists controller options for the H-Series including conventional industrial controller configurations and a compact DC controller intended to simplify integration with mobile robots such as AGVs and AMRs.

This can be useful when the application requires a robot arm to travel between workstations instead of remaining on one fixed pedestal.

However, mobile manipulation creates additional engineering requirements:

• base localisation;
• repeatable docking;
• power management;
• safety-system integration;
• robot calibration after movement;
• payload stability;
• networking.

Tool I/O

The H2017 provides 24 V DC tool power up to 3 A and digital I/O at the tool interface.

This supports direct integration of many:

• electric grippers;
• vacuum valves;
• part sensors;
• tool changers;
• process switches.

Check the exact electrical requirements of the selected tool. High-power devices may require external power.

Gripper selection

The best gripper depends on the application.

Vacuum tooling is often attractive for cartons, sheets, bags and flat objects.

Parallel electric grippers work well for repeatable components with defined gripping surfaces.

Pneumatic grippers can provide high force and simple operation where compressed air is available.

Custom tooling becomes necessary for complex, unstable or unusually large products.

The tool should be chosen before finalising the robot.

Do not forget payload used by the end effector

A large palletizing vacuum frame can consume several kilograms of H2017 payload.

If the tool weighs 6 kg, only 14 kg remains before reaching the nominal 20 kg system payload—and centre-of-gravity limits may reduce the practical margin further.

This is one reason to compare the H2017 with the 25 kg H2515 when handling products close to the limit.

Best Applications for the Doosan H2017

1. Machine tending

One of the strongest H2017 applications.

The long reach can allow one robot to access:

• CNC doors;
• part trays;
• inspection positions;
• infeed and outfeed conveyors;
• multiple machines where the layout allows it.

The 20 kg payload also leaves useful capacity for metal workpieces and a substantial gripper.

The buyer should confirm:

• door automation;
• machine I/O;
• chip and coolant exposure;
• part orientation;
• required cycle time.

IP54 protection may matter in particularly aggressive machining environments.

2. Palletizing

The combination of 20 kg payload and 1.7 m reach makes the H2017 a logical palletizing candidate for medium-weight cases and products.

It becomes less attractive when:

• box plus gripper mass approaches 20 kg;
• very tall pallets require an external lift;
• extreme palletizing throughput is required;
• 30 kg payload is needed.

Doosan’s newer P3020 should be evaluated for applications dominated almost entirely by palletizing.

3. Heavy material handling

The H2017 can move components between:

• conveyors;
• fixtures;
• machines;
• inspection stations;
• assembly positions.

This can remove repetitive lifting from operators while keeping the cell more flexible than fixed hard automation.

4. Welding

Doosan markets the H2017 for welding applications.

Its reach is useful for larger parts and fixtures, while six-axis articulation allows the torch to maintain orientation along complex paths.

A welding system still requires:

• welding power source;
• torch;
• wire feeder where applicable;
• fume management;
• fixture design;
• safety guarding or screens;
• process programming.

The robot being collaborative does not make the arc-welding process itself collaboratively safe.

5. Pick-and-place

The H2017 can perform conventional pick-and-place where products are too heavy or positions are too far apart for smaller cobots.

For light products and short distances, a smaller robot may deliver better economics and faster motion.

6. Assembly and component positioning

The joint torque sensors and force-control architecture can help when components need to be guided, aligned or inserted rather than simply positioned in free space.

The H2017 is particularly relevant when the parts themselves are heavy.

7. Filming and camera movement

Doosan lists filming as an H2017 application.

Long reach and repeatable six-axis movement can create controlled camera trajectories for studios, events and broadcast production.

This use case needs separate safety consideration because the payload can include expensive camera equipment positioned close to performers or crew.

8. Construction drilling and process tooling

Doosan also positions the H2017 for repetitive construction drilling.

The robot can potentially hold and position drilling equipment at defined locations, but the complete system must address:

• reaction forces;
• dust;
• tool vibration;
• workpiece positioning;
• IP54 environmental limits;
• mobile-base stability if the robot is relocated.

Is the Doosan H2017 Good for Palletizing?

Yes—especially when the product is below roughly 20 kg including tooling requirements and the 1,700 mm reach fits the pallet layout.

The H2017 was one of the earlier high-payload cobots to make heavier collaborative palletizing practical.

Its advantages include:

• 20 kg payload;
• long reach;
• force-sensing architecture;
• industrial controller ecosystem;
• Doosan palletizing software and turnkey experience.

But do a reach study first

A 1,700 mm maximum robot reach does not mean the H2017 can automatically service every point on every 1,700 mm-high pallet.

Palletizing geometry depends on:

• robot pedestal height;
• pallet dimensions;
• box height;
• tool length;
• required box orientation;
• joint limits;
• singularities;
• collision clearance.

Ask for a complete pallet pattern simulation.

When the P3020 may be better

Doosan’s current P3020 is specifically positioned for palletizing and offers:

• 30 kg payload;
• more than 2 m of published product-page reach;
• a five-axis architecture optimised for palletizing;
• higher capacity for heavier cases and tooling.

If the robot will spend its entire life stacking boxes, P3020 deserves serious consideration.

If the robot must also perform machine tending, welding, orientation-intensive handling or other general six-axis tasks, H2017 can remain the more flexible architecture.

When the Doosan H2017 Is Not the Right Robot

The H2017 is a strong robot, but several requirements point toward alternatives.

• Payload above 20 kg: consider H2515, P3020, UR30 or another heavier robot.
• Maximum production speed: newer high-performance cobots or conventional industrial robots may deliver better cycle times.
• Washdown or wet processing: IP54 is not equivalent to IP65 or IP67.
• Ceiling mounting: current Doosan H-Series documentation specifies floor-only installation.
• Very high positioning precision: ±0.1 mm may be insufficient for the process.
• Very small workcells: the 1.7 m reach and 79 kg arm may be unnecessarily large.
• Low-payload applications: a smaller Doosan M-Series or A-Series robot may cost less and move faster.
• Dedicated high-pallet applications: Doosan’s P3020 can provide a more purpose-built solution.
• Frequent robot relocation by people: a 79 kg arm is not a portable tabletop cobot.

The IP54 limitation matters

IP54 provides protection against limited dust ingress and splashing water, but it should not be interpreted as washdown protection.

Compare that with current alternatives such as:

• Universal Robots UR20 — IP65 robot arm;
• FANUC CRX-20iA/L — IP67 body.

For coolant-heavy machining, food production, outdoor work or wet cleaning, environmental protection can matter more than a small difference in payload or reach.

The floor-only mounting limitation matters

Many modern cobots can be installed:

• on the floor;
• on a wall;
• at an angle;
• inverted from a ceiling or frame.

The current H2017 documentation specifies floor-only mounting.

That may not affect a conventional pedestal cell, but it can eliminate the H2017 from applications where overhead mounting is essential to preserve floor space.

Doosan H2017 vs H2515, P3020, UR20 and FANUC CRX-20iA/L

There is no universal best heavy-payload cobot. The important differences are now large enough that the application should determine the shortlist.

H2017 vs UR20

This is one of the most relevant comparisons.

The two robots occupy a similar application envelope:

• approximately 20 kg standard payload;
• around 1.7 m reach;
• six-axis collaborative architecture;
• machine tending;
• welding;
• palletizing;
• material handling.

The H2017’s strongest differentiator is its torque sensor on every joint and Doosan’s force-control architecture.

The current UR20 counters with:

• up to 25 kg payload under specified conditions;
• 1,750 mm reach;
• 64 kg arm weight;
• IP65 protection;
• substantially higher published maximum TCP speed;
• the large Universal Robots software and integrator ecosystem.

Read our Universal Robots UR20 review before making the final choice.

H2017 vs FANUC CRX-20iA/L

The FANUC provides the same nominal 20 kg payload but only 1,418 mm reach.

In exchange, FANUC publishes:

• approximately 41 kg mechanical weight;
• ±0.04 mm repeatability;
• IP67 protection;
• floor, angle and inverted mounting;
• its established industrial service ecosystem.

Choose H2017 when the additional reach is essential.

Choose CRX-20iA/L when protection, mounting flexibility, lower arm mass or tighter repeatability is more important.

H2017 vs H2515

This decision is simpler:

• H2017: 200 mm more reach.
• H2515: 5 kg more payload.

Everything else should be compared at cell level.

H2017 vs P3020

Choose P3020 for heavy, dedicated palletizing.

Choose H2017 for a general-purpose six-axis application where end-effector orientation and broader task flexibility matter.

Use the Anton Robots comparison tool to compare available models.

Is the Doosan H2017 Worth It?

Yes, when the application genuinely uses its 1,700 mm reach, 20 kg payload and force-sensing architecture.

It becomes less compelling when those capabilities are not needed.

Where the value comes from

• Ability to automate heavier manual handling.
• Long reach without immediately moving to a large conventional robot.
• Six joint torque sensors.
• Force and compliance-control capability.
• Established safety architecture.
• Dart-Suite programming environment.
• ROS 2 integration for advanced users.
• Global Doosan dealer network.
• Compatibility with a broad range of industrial applications.

Where buyers can overpay

• Buying a 20 kg cobot for a 5 kg task.
• Paying for 1.7 m reach when every point is within 800 mm.
• Choosing a cobot when a faster fenced industrial robot would produce a better ROI.
• Designing around H2017 before comparing newer UR, FANUC and Doosan models.
• Buying the robot before selecting the end effector.
• Ignoring integration and safety costs.

A simple value test

Before buying, complete this sentence:

We need the H2017 because the application requires approximately _____ kg payload at _____ mm reach, and the robot must perform _____ while meeting a cycle time of _____ seconds.

If those numbers have not been defined, the project is not ready for robot selection.

Doosan H2017 Buying Checklist

1. Define the workpiece. Record weight, dimensions, material, orientation and fragility.
2. Define the tool. Include gripper mass, adapter, valves, cables and sensors.
3. Calculate total payload. Do not use workpiece weight alone.
4. Calculate centre of gravity. Confirm wrist moment and inertia requirements.
5. Map every robot position. Verify 1,700 mm reach through simulation.
6. Check mounting. The current H2017 specification is floor-only.
7. Check environment. Confirm whether IP54 is adequate.
8. Define cycle time. Measure the complete production cycle, not maximum robot speed.
9. Define accuracy. Confirm ±0.1 mm is sufficient.
10. Select safety architecture. Decide whether the cell needs scanners, guarding, interlocks or collaborative zones.
11. Define machine communication. List PLC, CNC, conveyor and sensor interfaces.
12. Select software. Decide whether Dart-Suite, standard pendant programming, API development or ROS 2 is required.
13. Review controller options. Confirm the exact controller included in the quote.
14. Request a reach simulation. Include full tooling and surrounding equipment.
15. Request a payload validation. Use actual centre of gravity and motion profile.
16. Request a safety assessment. Include the tool and workpiece, not only the arm.
17. Confirm warranty. Doosan currently publishes a one-year warranty starting from installation.
18. Confirm local service. Identify the responsible dealer or service centre before purchase.
19. Define acceptance criteria. Specify cycle time, availability, quality and fault-recovery requirements.
20. Compare alternatives. At minimum, compare H2515, P3020, UR20 and a relevant FANUC CRX model.

Pro tip: send the integrator a CAD model of the cell, the real workpiece, gripper concept and required cycle before requesting the final robot recommendation. A payload-and-reach shortlist is only the first stage of engineering.

How to Buy the Doosan H2017

The H2017 is generally purchased through Doosan Robotics or an authorised dealer/integrator rather than through a fixed-price global online store.

A useful enquiry should include:

• company and country;
• industry;
• application;
• workpiece weight;
• workpiece dimensions;
• required reach;
• cycle time;
• production hours per day;
• environmental conditions;
• existing PLC or machine interface;
• preferred gripper or process tool;
• safety requirements;
• target installation date.

Before placing the order, request:

• exact robot model and hardware revision;
• controller model;
• teach pendant model;
• complete bill of materials;
• payload and centre-of-gravity validation;
• reach simulation;
• cycle-time estimate;
• safety concept;
• software versions;
• communication interfaces;
• delivery lead time;
• warranty terms;
• training scope;
• commissioning scope;
• local support contact;
• acceptance-test procedure.

View the Doosan H2017 at Anton Robots or browse the full Doosan Robotics range.

If the application has not yet been narrowed to one robot, use Find My Robot before committing to the H2017.

What Is New for the Doosan H2017 in 2026?

The H2017 remains an active Doosan product

The most important finding is that the H2017 has not disappeared from the range.

Doosan still lists it as a current H-Series model alongside the H2515.

That matters because older industrial robots sometimes remain visible through distributor inventory even after manufacturers have shifted to a replacement generation. The H2017 is still part of Doosan’s current official lineup.

The market around it has changed

The H2017’s original 20 kg / 1,700 mm combination was unusually strong.

In 2026 the competitive landscape is harder.

Universal Robots now offers the UR20 with a very similar reach and an increased payload capability under specified conditions, while FANUC’s CRX range offers strong environmental protection and mounting flexibility.

The H2017 therefore has to be evaluated on the complete package—torque sensing, force control, Doosan software, integration and application fit—not on payload and reach alone.

P3020 changes the Doosan palletizing decision

Doosan’s P3020 now occupies the dedicated high-payload palletizing position above the H-Series.

It offers 30 kg payload and a working envelope above 2 m on Doosan’s current product page.

This means the H2017 should no longer be selected automatically for every Doosan palletizing project.

Its advantage is broader six-axis flexibility.

ROS 2 is the current development path

Doosan continues to maintain its official ROS 2 repository, including support for real and virtual operation, MoveIt 2 and simulation workflows.

The legacy ROS 1 repository now explicitly directs new users toward ROS 2.

The software ecosystem matters more than before

Dart-Suite has shifted Doosan’s proposition from simply selling robot arms toward a broader software-driven automation ecosystem.

For buyers with frequent process changes, this can matter as much as a small difference in mechanical specification.

No major H2017 hardware refresh found

We did not find evidence in Doosan’s current official documentation of a new 2026 H2017 generation with a fundamentally different payload, reach or mechanical architecture.

The current buying decision should therefore be based on the established H2017 platform and the latest controller/software ecosystem—not an assumed next-generation H2017.

Doosan H2017 FAQ

What is the payload of the Doosan H2017?

The H2017 has a published maximum payload of 20 kg.

Remember that this includes the end effector, adapters, sensors, cabling and the workpiece.

What is the reach of the H2017?

Doosan publishes a maximum working radius of 1,700 mm.

How accurate is the Doosan H2017?

Its published position repeatability is ±0.1 mm according to ISO 9283.

How fast is the H2017?

Current Doosan H-Series documentation lists a TCP speed around 1 m/s or above 1 m/s depending on the specific document revision. Actual application speed depends on payload, path, safety configuration and process requirements.

How many axes does the H2017 have?

Six rotational axes.

Does the H2017 have torque sensors?

Yes. Doosan states that the H2017 uses a high-performance torque sensor on each of its six joints.

Can the Doosan H2017 detect collisions?

Yes. Collision detection is part of Doosan’s safety-monitoring architecture and uses joint torque information.

The appropriate sensitivity and stop behaviour must be configured for the application.

Is the Doosan H2017 a collaborative robot?

Yes.

However, a collaborative robot arm does not automatically make every completed application safe for unrestricted human access. The complete cell requires risk assessment.

Can the H2017 operate without safety fencing?

Potentially, depending on the tool, payload, speed, workpiece and risk assessment.

Some H2017 applications will still require guarding or other protective measures.

Is the H2017 waterproof?

No waterproof claim should be assumed.

The current H2017 specification is IP54.

Can the H2017 be mounted on a ceiling?

The current H-Series documentation specifies floor-only installation for H2017.

How much does the H2017 weigh?

Current V3 documentation lists approximately 79 kg.

Older Doosan documents can show lower figures, so buyers of used equipment should check the exact hardware revision.

What temperature can the H2017 operate in?

Doosan specifies an operating range of 0°C to 45°C.

Is the H2017 good for palletizing?

Yes. Its 20 kg payload and 1,700 mm reach make it suitable for many palletizing applications.

For heavier products or specialised high-pallet work, compare Doosan’s P3020 and H2515.

Can the H2017 tend CNC machines?

Yes. Machine tending is one of Doosan’s primary applications for the model.

The long reach can be particularly useful for machine access and multi-station layouts.

Can the H2017 weld?

Yes. Doosan lists welding among the H2017’s applications.

The complete system requires welding equipment, fixtures and appropriate safety controls.

Does the H2017 support force control?

The H-Series uses built-in joint torque sensors and supports force/compliance-control functionality.

Validate the exact control mode and required process performance for the application.

Does the H2017 support ROS?

Yes. Doosan maintains an official ROS 2 stack covering its robot lineup.

Does the H2017 support ROS 1?

Legacy ROS 1 resources exist, but Doosan states that ROS 1 support ended in May 2025 and directs users toward ROS 2.

Can I simulate the H2017?

Yes. Doosan’s ROS 2 ecosystem includes virtual operation and simulation resources, and integration workflows are available for tools such as Gazebo and MoveIt 2.

What software does the H2017 use?

Doosan’s current ecosystem includes its robot controller software, teach-pendant workflow and Dart-Suite platform, with APIs and ROS 2 available for more advanced development.

What is the H2017 warranty?

Doosan’s current published warranty period is one year from installation, subject to its warranty terms and exclusions.

How long is a Doosan collaborative robot designed to last?

Doosan’s service FAQ states that its robots are designed around a maximum life of approximately 35,000 operating hours, while noting that actual life depends on operating conditions.

This is a design estimate, not a guaranteed production lifetime.

What is the difference between H2017 and H2515?

H2017 provides 20 kg payload and 1,700 mm reach.

H2515 provides 25 kg payload and 1,500 mm reach.

Choose between reach and payload.

What is the best alternative to the H2017?

For a similar long-reach heavy cobot, the Universal Robots UR20 is one of the closest comparisons.

FANUC CRX-20iA/L is attractive when IP67 protection, lower robot weight, flexible mounting or tighter repeatability are more important.

Within Doosan, compare H2515 for additional payload and P3020 for dedicated palletizing.

Is the Doosan H2017 better than the UR20?

Not universally.

H2017 offers six joint torque sensors and Doosan’s force-control and safety architecture.

UR20 currently provides similar reach, higher maximum payload capability under defined conditions, IP65 protection, lower arm weight and substantially faster published maximum TCP performance.

The better robot depends on the application and total cell proposal.

Is the Doosan H2017 still current in 2026?

Yes. Doosan Robotics continues to list the H2017 as a current H-Series product.

Is the Doosan H2017 worth buying?

Yes, when the application uses its long reach, heavy payload and force-sensing capabilities.

It is less attractive if a smaller, faster or more environmentally protected robot can perform the same task for lower total cost.

Final Verdict: Should You Buy the Doosan H2017?

Shortlist the Doosan H2017 when you need a genuine heavy-payload collaborative robot with 20 kg capacity, 1,700 mm reach and force-aware six-axis control.

Its strongest use cases are not small collaborative tasks. They are applications where reach and payload start to push beyond the practical limits of conventional lightweight cobots: machine tending, palletizing, metal handling, welding and long-distance material transfer.

The six joint torque sensors remain a meaningful differentiator. Combined with Doosan’s safety functions, force control and Dart software ecosystem, they create a capable foundation for flexible industrial automation.

But the market has moved.

The UR20 now competes extremely closely on reach while offering stronger environmental protection and faster published motion. FANUC provides lighter, highly protected CRX alternatives. Doosan’s own H2515 offers another 5 kg of capacity, and P3020 is a more specialised choice for heavy palletizing.

That means the H2017 should not win because someone recognises the model name.

It should win because the application needs its exact combination of:

• 20 kg payload;
• 1,700 mm reach;
• six-axis articulation;
• joint torque sensing;
• Doosan force control;
• Doosan software and integration.

Before ordering, calculate the real payload with tooling, model the complete reach envelope, confirm IP54 is sufficient, design the safety system and compare the quoted cycle time against at least two alternatives.

If the H2017 still wins after that engineering exercise, it remains a very credible industrial cobot in 2026.

View the Doosan H2017 at Anton Robots, browse other collaborative robots, or contact Anton Robots to compare models and supplier options.