Ocean Engineering 291 (2024) 116282

Contents lists available at ScienceDirect

Ocean Engineering
journal homepage: www.elsevier.com/locate/oceaneng

Assessing damage and predicting future risks: A study of the Schilling Titan
4 manipulator on work class ROVs in offshore oil and gas industry
Dalibor Ivanega , Marcin Szczepanek *
Faculty of Marine Engineering, Maritime University of Szczecin, ul. Waly Chrobrego 1-2, 70-500, Szczecin, Poland

A R T I C L E I N F O A B S T R A C T

Handling Editor: Prof. A.I. Incecik This article undertakes a comprehensive investigation of the damage incurred by the Schilling-manufactured
Titan 4 seven-function manipulator during the period of 2017–2022. This particular manipulator stands as the
Keywords: most widely employed equipment on work class remotely operated vehicles (ROVs) globally. The article delves
Manipulator into a detailed discussion of ROVs in general, with a specific focus on those belonging to the work class category.
ROV
Furthermore, the manipulator is dissected into its constituent elements, as delineated by the manufacturer, and
Offshore
each individual component is subjected to separate evaluation. The damages that were encountered are
Oil and gas
Subsea described in detail, along with the corresponding repair processes.
Schilling To quantify the likelihood of future damage to the various manipulator elements over the subsequent five-year
Robotics period, the obtained damage results are inputted into the Geogebra program. This enables the calculation of the
probability of damage occurrence for each individual component.
The research findings are ultimately summarized and juxtaposed with the real-life experiences of ROV op­
erators, which are captured through a meticulously designed questionnaire. The questionnaire is carefully
constructed to elicit valuable insights from the operators regarding the most common faults or defects observed
in the Titan 4 manipulator.
It is important to note that this manipulator-focused investigation is based on the author’s first-hand expe­
rience as a ROV Supervisor in the offshore oil and gas industry. The combination of practical expertise and the
empirical data gathered from the questionnaire responses ensures a comprehensive and well-rounded exami­
nation of the Titan 4 manipulator and its associated operational challenges.

1. Introduction the offshore oil industry. Since then, more than 90% of ROVs produced
have been tailored for commercial offshore operations, including sup­
Remotely operated vehicles (ROVs) are underwater vehicles port for oil and gas drilling, as well as inspection, burial, and repair of
controlled remotely from the surface. They are connected to a surface pipelines and telecommunications cables (Cochran, 2019; Nichols and
platform through a cable, which supplies power and enables commu­ Williams, 2009).
nication for controlling the vehicle. There are three primary types of With the increasing depths of oil exploration and production, the
ROVs: free-swimming tethered vehicles, bottom-crawling tethered ve­ commercial ROV industry has faced the challenge of keeping up. Current
hicles, and towed vehicles. The most commonly used type is the free- exploration depths now reach up to 3000 m. The remaining ROVs serve
swimming vehicle, equipped with thrusters for manoeuvring in three military and scientific purposes. Military applications involve subma­
directions. It features onboard video cameras that provide visual feed­ rine rescue, mapping, reconnaissance, recovery, and mine counter­
back, making it suitable for mid-water or bottom observation and measures. Scientific applications span diverse fields such as biology,
intervention (Cochran, 2019). physics, geology, and chemistry. The depths for scientific work vary
The offshore community witnessed the introduction of remotely from a few meters to 10,000 m (Cochran, 2019; Kennish, 2019).
operated vehicles (ROVs) in 1953. Over the following years, additional The demand for work class remotely operated vehicles (WCROVs)
vehicles were constructed to meet military and government research has traditionally been influenced by the requirements of the global
needs. In 1975, the first commercial ROV was developed specifically for offshore oil and gas industry. This trend is expected to persist in the near

* Corresponding author.
E-mail address: [email protected] (M. Szczepanek).

https://doi.org/10.1016/j.oceaneng.2023.116282
Received 27 July 2023; Received in revised form 25 October 2023; Accepted 4 November 2023
Available online 25 November 2023
0029-8018/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

to medium term. However, a new market segment is emerging, namely instances where they are positioned at the bottom (Ribas et al., 2011;
wind farms. The growth of wind farms is regarded as a crucial catalyst Sánchez et al., 2020; Simetti et al., 2017).
for northern hemisphere countries with coastal access and favourable There are review papers available for a more detailed overview of
wind conditions to attain zero emissions (Thompson and Knight; Teague underwater manipulators, discussing their design, modelling, control,
et al., 2018). and theoretical background (Sivčev et al., 2018a; Bogue, 2015; Petillot
According to Archer Knight’s analysis, the offshore wind farms sector et al., 2019; Antonelli, 2014).
presently constitutes approximately 29% of the global demand for Nowadays a vast majority of manipulators used subsea on ROVs are
ROVs. Nonetheless, the oil and gas sector continue to dominate with operated by humans – ROV Pilots and Co-Pilots respectively. Human
approximately 64% of the demand, remaining the primary driver of pilots teleoperate the manipulators based on visual feedback, but limited
ROV market. The remaining portion of the demand is attributed to the perception and visibility can lead to collisions and significant damage.
defence and research sectors (Thompson and Knight). Graphics pre­ Which is discussed later in this article. In (Sivčev et al., 2018b), the
sentation on Fig. 1. ROV Demand share. collision detection mechanism is integrated into a commercial ROV
Work Class ROVs (WCROVs), are advanced intervention vehicles manipulator control system and evaluated through simulations and ex­
designed to handle demanding operations in the underwater oil and gas periments using a real industry-standard underwater manipulator. The
industry, such as inspection, maintenance, and system repair (IMR), as presented collision sensing solution has the potential to assist pilots,
well as survey tasks. WCROVs encompass a comprehensive range of reducing task load, operational time, and costs in subsea inspection,
capabilities required for projects involving underwater work. These repair, and maintenance operations.
submersible vehicles are equipped with plug and play systems, enabling Fig. 2 illustrates a Work Class ROV (WCROV) equipped with a Rig­
easy interchangeability of tooling and tools. Additionally, they utilize master manipulator on the port side and a Titan 4 manipulator on the
the StationKeep positioning system, which ensures a highly stable starboard side of the vehicle. This configuration enhances the ROV’s
environment for intervention tasks (Christ et al., 2014; McLean et al., capability to handle demanding underwater operations in the oil and gas
2020). industry. In Fig. 3, a different configuration is depicted, showcasing a
Underwater manipulators, such as robot arms, are commonly used in WCROV with an Atlas manipulator on the port side and a Titan 4
subsea intervention operations performed by unmanned underwater manipulator on the starboard side of the vehicle. This arrangement
vehicles (UUVs) like remotely operated vehicles (ROVs) and autono­ provides similar functionalities, allowing the ROV to perform intricate
mous underwater vehicles (AUVs). These manipulators, also known as tasks effectively. The Atlas manipulator, similar to the Titan 4, possesses
Underwater Vehicle Manipulator Systems (UVMS), typically resemble six degrees of freedom (DOF) along with a grip. In contrast, the Rig­
human arms and consist of interconnected rigid bodies with revolute master manipulator has a total of four degrees of freedom, supplemented
joints and end-effectors like grippers or tools. They are often equipped by a grip (Sivčev et al., 2018a).
with cameras and spotlights for observation purposes (Sivčev et al., Examples in which ROVs are typically working:
2018a; Huang et al., 2020; Aldhaheri et al., 2022).
Underwater manipulators find applications in various industries, • Depth ranges below 200m is an average temperature 4 deg Celsius.
including offshore oil and gas, marine renewable energy, marine civil • None of the examinate ROV systems in this article was working
engineering, marine science, and military sectors (Capocci et al., 2017; below 2500m.
Dalhatu et al., 2023). They are designed for different purposes, such as • Temperature on deck in tropical areas is around 40 deg Celsius.
lifting heavy objects, attaching detachable grippers to sunken objects,
fixing underwater vehicles to structures or walls, inspection tasks, These figures exemplify the versatility and adaptability of WCROVs,
dexterous intervention operations, and more. Work class ROVs typically
have two manipulators, with one serving to hold the ROV near the
structure while the other performs the actual intervention task (Sivčev
et al., 2018a).
The tasks executed by underwater manipulators include pipe in­
spection, salvage operations, mine disposal, surface cleaning, valve
operation, drilling, rope cutting, cable laying and repair, debris and
fishing net removal, biological and geological sampling, archaeological
work, and more (Christ et al., 2014; Fletcher, 2000; Davey et al., 1999;
Mazzeo et al., 2022; Coleman et al., 2003). While manipulators are
usually located at the front of the underwater vehicle, there are

Fig. 2. WCROV equipped with Rigmaster (Port side) as well as Titan 4 (Star­
Fig. 1. ROV Demand share (Thompson and Knight). board side) [author’s photos].

2
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

Fig. 3. WCROV equipped with Atlas (Port side) as well as Titan 4 (Starboard
side) [author’s photos].

as they can be equipped with different manipulator systems based on the

specific requirements of the underwater projects.
It is worth noting that manipulators come in various sizes and con­
figurations. The manipulator being examined in this study is the Schil­
ling Robotics Titan 4 manipulator. It is a unique piece of equipment that
is most commonly used on work class ROVs worldwide. Consequently,
the Schilling T4 manipulator (Fig. 4.) currently faces no direct compe­
tition. Statement is based on a authors own professional experience from
ROV industry as well as from a review paper (Sivčev et al., 2018a) where
Fig. 4. Schilling Titan 4 manipulator [author’s photos].
can be seen thorough analysis of nowadays manipulators used subsea in
offshore industry.
components, over a five-year period.
2. Materials and methods The research problem at hand involves estimating the most
frequently recurring damage to individual elements of the manipulator.
The Titan T4 manipulator represents a highly advanced device that Furthermore, it entails analysing the reasons behind these damages and
encompasses multiple disciplines, including mechatronics, robotics, proposing preventive measures to mitigate them effectively.
hydraulics, and mechanics. Constructed from titanium and diffusion The answers to these questions are subsequently summarized and
welded, it boasts corrosion resistance as one of its key attributes. The compared with the actual damages recorded during the study period.
Titan 4 is a high velocity, compact and lightweight, spatially corre­ During the research conducted over a five-year period (2017–2022),
spondent manipulator system specifically designed for remote manipu­ the Titan 4 manipulator, manufactured by Schilling Robotics, was uti­
lation. It has six DOF plus a grip and is controlled by a master controller. lized in various offshore subsea projects. These projects encompassed a
Operator movements introduced at the master control arm are dupli­ range of activities including heavy construction, Inspection Mainte­
cated by the slave arm while maintaining the spatial correspondence nance and Repair (IMR), cable laying, CP survey, rigid pipe laying,
between master and slave (O’Donnell, 1995; Harden et al., 2009; Bird decommissioning, among others. Observations were made on different
et al., 2018). The master and slave arms are show on Fig. 5 as a block offshore vessels operating in various locations worldwide. The aim was
schematic diagram to represent how a typical hydraulic manipulator is to gather objective data to provide an accurate representation of the
incorporated in ROV. manipulator’s performance.
As interesting facts, besides a human operator experience and skills, The collected data is presented in Table 1, accompanied by concise
presents Fig. 6, which represents factors affecting underwater manipu­ descriptions of the observed faults at the time and the subsequent re­
lator performance such as mechanical design, kinematics, dynamics, pairs. It is important to note that the descriptions of damage and repair
actuators, etc (Sivčev et al., 2018a). procedures are articulated in the specialized language commonly used in
The primary components of the manipulator Schilling Titan 4 Fig. 7 the ROV industry. Consequently, certain expressions may pose difficulty
include the azimuth, shoulder, upper arm, elbow, forearm, pitch & yaw, in comprehension. To maintain authenticity, all descriptions were
wrist, and jaw. Throughout this article, has been examined failures of derived from ROV logbooks.
these components, as well as the compensator, camera, and master arm The Titan 4 manipulator from Schilling Robotics comprises several

3
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

Fig. 5. Long-line power and telemetry connections of manipulator Titan 4 (TechnipFMC).

Fig. 6. Factors affecting underwater manipulator performance (Sivčev et al., 2018a).

Fig. 7. Manipulator Titan 4 main components (Guo and Tian, 2022),

4
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

Table 1 Table 1 (continued )

Defects in the basic elements of the Titan 4 manipulator encountered during 1. Azimuth
testing (own study).
Number of Damage type Repair description
1. Azimuth damages
Number of Damage type Repair description of the sliprings and re-
damages assembled. Arm tested on deck
1 Found a hair crack on a female Connection replaced for 2 h - no erratic movement
JIC connection observed and no noise or heat as
before. Slave arm to be tested
2. Upper arm subsea when project time allows
Number of Damage type Repair description
damages 5. Forearm
Number of Damage type Repair description
0 – – damages
3. Shoulder actuator 1 Internal leak Seals replaced
Number of Damage type Repair description 2 Leaking Fault found to be the 16 3/8
damages fastners holding the plate down
1 Shear pin broken Changed shear pin and piston that the seanet connector is on
rod were less than finger tight
2 Leaking linear actuator Seals replaced 6. Pitch and Yaw Actuators
3 Leaking linear actuator Seals replaced Number of Damage type Repair description
4 Loosing oil Found punctured hose – replaced damages
5 Leaking linear actuator Seals replaced
6 T4 uncontrolled shoulder After checking slave diagnosis on 1 Leaking Seals replaced
movement when in up position MCU found shoulder position 2 Leaking Found wrong type of bolts fitted,
reading erratic, inspected cleaned up the joint replaced the
shoulder resolver, found broken bolts with the correct ones, all
wire from previous solder fix. seals on slip ring and bolts
Replaced resolver, fill & bleed replaced. Leak still present at
arm pitch joint. Replaced with spare
7 Piston shear pin broke replaced sheer pin and bolts forearm. Stripped both outer
8 Leaking linear actuator Seals replaced pitch/yaw joints and inspected
9 Found delamination between Resolver replaced closer. Noticed outer wear rings
inner mounting ring and inner very worn. Showing wear on
coils on Shoulder resolver outer slipring plates. Changed all
10 Leaking linear actuator Seals replaced slipring seals and re-installed
11 Leaking linear actuator Seals replaced 3 Leaking Pitch joint allen bolts retorqued
12 Shear pin sheared during subsea Shear pin replaced 4 Leaking Seals replaced + slip ring face
ops replaced
13 Shear pin sheared during subsea Shear pin replaced 5 Erratic Pitch control Fault was traced to the master
ops controller. Unit to be sent back
14 Shear pin broken Changed shear pin and piston for repair
rod 6 Leaking Seals replaced
7 Erratic Pitch control Fault was traced to the master
4. Elbow actuator controller. Unit to be sent back
Number of Damage type Repair description for repair
damages 8 Leaking Seals replaced
1 Leaking Seals replaced 9 Yaw function stopped working Changed MOOG valve
2 Leaking Seals replaced 7. Wrist Actuator
3 Leaking Seals replaced Number of Damage type Repair description
4 Leaking Seals replaced damages
5 Leaking Seals replaced
6 Leaking Seals replaced 1 Slow and unresponsive Moog valve changed
7 Sluggish movement & being Replaced elbow MOOG valve operation
unable to lift the class 4 TT in 2 Unresponsive Investigation found that an O
water Ring on the Oil Supply Bolt
8 Elbow not responding Investigation showed that the O- needed changing on the wrist
Rings (10 x 004–0627) were actuator
doubled up on the Slip Rings, 3 Wrist uncontrolled movement Fault traced to main PCB –
and bots loose (20 x 002-0709P). Replaced
Addtionally it was observered 4 wrist creeped during dive and even with top side switched off.
that the resolver wires had been on deck Replaced Moog Servo for wrist
damaged due to the clamping joint tested all good. Suspect 111
mechanism. Resolver was contaminated as comp oil was
changed out heavily contaminated with 111,
9 Leaking Seals replaced so hyd circuit probably also
10 Erratic movement Removed lower arm and stripped contaminated. Wrist creep still
the elbow joint. Cleaned and noted a day later see 24/12/13.
inspected the joint with no Going to test Moog servo on
defects found. Re- torqued the another arm on stand to confirm
allen grub screws on elbow joint T4 (S/N: 104697, SAP?) slave
to 6IN/LB. Slip rings inspected arm PCB replaced to stop wrist
and all seals good. found oil creep when in rate mode and
absorbent pad debris compressed arm frozen. Seems much better.
into 2 x oil ways - removed. Also the glitch in the arm when
Replaced the o rings on the end the LED light are turned off

(continued on next page)

5
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

Table 1 (continued ) Table 1 (continued )

1. Azimuth 1. Azimuth

Number of Damage type Repair description Number of Damage type Repair description
damages damages

seems to be gone. Going to test Removed this and fitted the

the old PCB in another arm to shoulder moog and check valve
confirm if its U/S Spare T4 (S/N from the spare arm. The fault
11005038) servo/controller PCB symptoms remained however
connectors cleaned and when the shoulder resolver was
reconnected. Function checked disconnected from the joint the
spare T4 all joints ok arm operated smoothly. Suspect
possibly faulty shoulder resolver.
8. Jaw and Tool Actuator
The resolver was removed. fitted
Number of Damage type a Repair description
shoulder resolver Further invest
damages
required on arm with the new
1 Leaking Piston Bearing and Piston shaft resolver fitted the arm still
found damaged – Replaced juddered with hydraulics applied
2 Jaw not closing Replaced both Bearings and T- and when brought live drove
Bar itself on to the deck. All these
3 Jaw not closing Replaced both Bearings and T- tests had been carried out using
Bar the spare master controller
4 Jaw not closing Replaced both Bearings and T- thought to be serviceable. H29’s
Bar master arm was connected to the
5 Jaw not closing Replaced both Bearings and T- arm and produced similar fault
Bar conditions, as a final check the
6 Fingers claw cracked Found double finger bushes master arm from herc 11 was
missing and bit broken off. connected and the fault was no
Replaced longer apparent
7 Jaw not closing Fingers claw changed 4 Showing „Fatal Telemetry Error Fault traced down to topside 232
8 Jaw not closing Replaced both Bearings and T- “ chip
Bar 5 Jaw stuck open intermittently Traced to Master Control Unit
9 Jaw not closing Replaced both Bearings and T- 6 Jaw open/close command not Replaced the conductive jaw
Bar work band on the Master controller
10 Two fingers broken Replaced both Bearings and T-
Bar
11. Leak on LVDT Seals replaced main components and elements, including the Azimuth, Upper arm,
12. Jaw not closing Replaced both Bearings and T- Shoulder actuator, Elbow actuator, Forearm, Pitch and Yaw actuators,
Bar
Wrist actuator, Jaw and Tool actuator, Compensator, Camera, and
13. T – Bar found split in half Replaced both Bearings and T-
Bar Master Controller. The faults encountered during the research are
14. Broken single finger T – Bar changed categorized based on the primary elements of the Titan 4 manipulator
15. Wear jaw bearing on single Single finger changed are presented in Table 1.
finger
16. Jaw not closing Bearing Replaced

9. Compensator 2.1. Assessment of occurring damages

Number of Damage type Repair description
damages The analysis of the damage descriptions presented in Table 1 enables
1 Leaking Three of its eight mounting bolts us to evaluate the frequency of failures and identify the elements that are
head broken most commonly damaged in the Titan 4 manipulator. This assessment
2 No comms Broken arm on Seanet connector
facilitated the classification of damages, which is presented in Table 2.
3 Leaking Damaged hose replaced
By examining the frequency of occurrence for each type of damage, it
10. Camera becomes possible to discern patterns and trends regarding the vulner­
Number of Damage type Repair description
able components of the manipulator. Such insights provide valuable
damages
information for maintenance and improvement strategies in the field of
1 Not working Short found on co-ax cable –
underwater operations.
cable resoldered
Table 2 showcases a comprehensive classification of damages based
11. Master Controller on the observed frequencies of failure incidents. This classification offers
Number of Damage type Repair description
damages
a structured overview of the elements within the Titan 4 manipulator
that are most susceptible to damage during offshore subsea projects. The
1 Blank screen LCD screen cable damaged
data presented in the table serves as a foundation for further analysis
2 Showing „ Slave arm continuity Found shoulder resolver being a
error “when slave arm fully bit dodgy. On opening the and decision-making processes related to optimizing the manipulator’s
down resolver housing we found it was performance and enhancing its resilience to potential faults.
full of gunk, we cleaned and Understanding the frequencies and patterns of damages is crucial for
flushed the arm thoroughly
before replacing the resolver
with a spare. Table 2
3 7 function judders erratically Substituted the slave arm PCB as Evaluation of damage classification.
when hydraulics is applied the there were so many different
Classification of frequency of failures
arm is sluggish and there is no faults, fault remaned replaced
full range of movement on the the PCB. The shoulder moog 0–5 Rarely
elbow joint valve was replaced with a
6–11 Occasionally
previously used one but now the
arm drove itself in the deck. 12 and more Frequently

6
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

mitigating risks, improving operational efficiency, and ensuring the load on the manipulator, particularly on the jaws. Consequently, the
reliability of the Titan 4 manipulator in future offshore subsea projects. operator’s experience becomes crucial, especially when working in
This information contributes to the development of strategies aimed at conditions with poor visibility, as determining the manipulator’s precise
minimizing downtime, reducing repair costs, and enhancing overall position for specific tasks, such as gripping, valve operation, or manip­
operational productivity. ulating objects subsea, can be challenging. As a result, certain manipu­
Table 3 provides a comprehensive summary and evaluation of the lator positions may occur unintentionally.
individual elements comprising the Titan 4 manipulator, based on the Several different tools and toolings are used subsea during ROV
frequency of damages encountered. This evaluation offers valuable in­ construction works, operated with Titan 4 manipulator. As an example:
sights into the vulnerability of each element, shedding light on the secateurs (Fig. 10) designed to cut thick and heavy slings that cannot
components that are most susceptible to failures during offshore subsea easily severed with a standard ROV knife. Hydraulic class 4 torque tool
operations. The table serves as a valuable reference for maintenance (Fig. 11) commonly used to operate valves on subsea structures.
planning and optimization strategies. Various tools are employed for underwater tasks, as exemplified in
Furthermore, Fig. 8 visually represents the number of failures Figs. 12 and 13, demonstrating the practical applications of the Titan 4
observed for each year through a bar graph. This graphical representa­ manipulator.
tion enables a quick and intuitive understanding of the annual distri­
bution of failures in the Titan 4 manipulator. The bar graph provides a b) Shoulder Actuator – 14 damages
visual narrative of the fluctuating trends and patterns of damages over
the specified time period, offering additional context to the numerical The shoulder actuator of the Titan 4 manipulator is a hydraulic
data presented in Table 3. actuator responsible for the vertical movement of the arm, allowing it to
The information presented in Table 3 and Fig. 8 collectively be lowered or raised. Fig. 14 displays the complete manipulator as­
contribute to a comprehensive understanding of the performance and sembly, with specific emphasis on the hydraulic actuator, as indicated.
reliability of the Titan 4 manipulator. By analysing the frequency and The presence of gray plastic material serves as a mechanical safeguard,
distribution of damages across the individual elements and over time, protecting the piston pin against potential damage.
stakeholders in the offshore subsea industry can make informed de­
cisions regarding maintenance strategies, spare part management, and c) Elbow actuator – 10 damages
potential design enhancements to improve the robustness and efficiency
of the manipulator system. Leaks are a commonly observed issue in hydraulic systems. The joints
of the Titan 4 manipulator are meticulously engineered, incorporating
multiple O-rings with varying diameters and thicknesses. As with any
2.2. Description of the most frequently damaged elements device that incorporates diverse types of O-rings, it is inevitable to
encounter leaks at some point. For instance, when an ROV is situated on
The manipulator components have been categorized into distinct the deck, the temperature can exceed 50 ◦ C under direct sunlight in
elements or groups in order to precisely classify the frequently damaged tropical regions. Subsequently, when the ROV is submerged to opera­
elements during operation. There are a total of eleven elements in the tional depths of over 1000 m and beyond, the environmental tempera­
manipulator system. Fig. 8 provides a clear identification of the elements ture in such conditions decreases to approximately 4 ◦ C. It is important
that are particularly vulnerable to damage. to note that the ambient pressure is not a significant factor in this sce­
Table 4 provides individual elements model numbers and part codes nario, as the entire internal space of the manipulator is filled with oil and
for damaged components. connected to a compensator, which maintains the pressure inside the
manipulator consistently at a level 7–10 psi (0.5–0.7 bar) higher than
a) Jaw and Tool Actuator - 16 damages the ambient pressure.
A noteworthy instance of a leak involves the compromised slip rings
The most frequently damaged components in the manipulator system within the elbow actuator of the manipulator. The damage occurred as a
are the jaws/tool actuator, which control the opening and closing of the result of mishandling cargo during an operation involving two collab­
jaws. Analysis of the data presented in Table 1 reveals that out of 16 orating underwater vehicles. A significant dynamic hydraulic impact
recorded failures, 15 were related to issues with jaw movement and took place at the midpoint of the joint. Interestingly, it was observable
damaged bearings and T-pieces. Fig. 9 displays a typical set of jaws with how the manipulator swiftly straightened itself from its previously
a detailed description of each individual element. curved position without any intervention from the operator. Subsequent
The jaws serve as the terminal element of the manipulator system. to this incident, the manipulator exhibited improper functioning within
The primary controller, known as the Master arm, utilized for under­ the arm cylinder domain, and the oil compensator of the manipulator
water manipulator control, lacks feedback mechanisms to indicate the

Table 3
Assessment of individual elements of the Titan 4 manipulator based on the frequency of failures (own study).

7
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

Fig. 8. Frequency of failures in the main elements of the T4 manipulator over a five-year period.

proximity subsequent to the disassembly process. Fig. 16 reveals the

Table 4
presence of evident damage, which had been suspected prior, as it is now
Model numbers and part codes for damaged components.
confirmed that one of the rings has incurred damage. This damage can
Component Part number be attributed to a momentary surge in pressure that ensued during the
Shaft, shear pin 001–5294 swift rectification of the manipulator, resulting in ring impairment and
End, shear pin 001–5259 subsequent occurrence of a leak.
Seal, Piston linear actuator 004–0649
O-ring,2–031,Buna 90 004–0669
Seal, Rod linear actuator 004–0650 d) Pitch & Yaw actuator – 9 damages
O-ring,.189X.039,Buna90 004–0627
O-ring,189X.039,N70(Apple) 004–0585 Similarly, to the Elbow Actuator, leaks are also the most commonly
O-ring,1.417X.039,N70(Apple) 004–0586 encountered issue in the case of the Pitch and Yaw actuator (Fig. 17).
Valve, servo, Moog 30–389B, Wlemo 006–2330
Bearing, Jaw actuator, T4 001–9835
The Pitch and Yaw actuator performs movements in two directions:
Pin,.5X2.0,Jaw, T2 001–1840 Pitch, which involves upward and downward motion, and Yaw, which
RSLR/LVDT/WR,T2/T3 101–1409 involves left and right motion. These are delicate movements used for
PCB, Master processor, MCU 005–5195 precisely positioning the manipulator before commencing a given task
or when using lightweight tools, which the manipulator handles with
great proficiency.
began to fill. These observations indicate the presence of a leak within
A noteworthy occurrence of damage involves the discovery of a
the manipulator. Specifically, the leak originates from the high-pressure
cracked actuator housing (Fig. 18) during routine checks conducted on
side, infiltrating into the low-pressure system or compensated space.
the ROV system. This particular incident captures attention due to its
In Fig. 15, the slip rings and arm cylinder are observed in close

Fig. 9. Jaw elements [author’s photos].

8
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

Fig. 13. Connection inspection [author’s photos].

Fig. 10. Secateurs [author’s photos].

Fig. 14. Shoulder actuator on the Titan 4 manipulator [author’s photos].

Fig. 11. Class 4 Torque tool [author’s photos].

Fig. 15. Slip rings and Elbow actuator [author’s photos].

infrequent nature, as the author, with a substantial five-year span of

professional experience, has encountered this specific type of damage on
just a single occasion. The occurrence of such a crack in the actuator
housing is an infrequent event, highlighting its atypical nature within
Fig. 12. Shackle disconnection [author’s photos]. the context of regular operations.

9
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

It can be inferred that the jaw bearings, constructed from a softer

material like brass, are intentionally designed to sustain damage in order
to protect the more expensive components of the manipulator. While
this aspect may initially appear as an area for improvement due to
frequent destruction, it is actually a deliberate strategy employed by the
manufacturer of the Titan 4 manipulator.

b) Shoulder actuator

The Titan 4 manipulator lacks a feedback mechanism between the

manipulator itself and the controller utilized by the operator. Conse­
quently, determining the weight of underwater objects that require
lifting becomes challenging. As mentioned earlier, the operator’s expe­
Fig. 16. Detail of damaged ring [author’s photos].
rience becomes crucial in such situations.

c) The system’s most vulnerable component is the seal, which is prone

to leakage when damaged.Elbow Actuator

Leaks are a common issue in hydraulic systems, including the Titan 4

manipulator. The manipulator’s joints are designed with multiple O-
rings of different sizes and thicknesses, making occasional leaks inevi­
table. Environmental conditions, such as extreme temperature varia­
tions, can contribute to the occurrence of leaks. In the case of the
compromised slip rings in the manipulator’s elbow actuator, the damage
occurred due to mishandling during an operation involving underwater
vehicles. A dynamic hydraulic impact caused the damage, leading to
Fig. 17. Pitch & Yaw actuator [author’s photos]. swift self-straightening of the manipulator. Subsequently, improper
functioning within the arm cylinder and the filling of the oil compen­
sator were observed, indicating a leak within the manipulator. Fig. 15
displays the slip rings and arm cylinder after disassembly, while Fig. 16
confirms the suspected damage with visible impairment to one of the
rings. This damage can be attributed to a momentary surge in pressure
during the rapid rectification of the manipulator, resulting in a leak.

d) Pitch & Yaw actuator

The occurrence of a cracked actuator housing is highly unpredictable

and challenging to anticipate. This type of damage is not attributed to
operator error. Ideally, it is recommended to detect any signs of cracking
prior to commencing operations, specifically before submerging the
ROV into water, in order to prevent accidental oil leakage into the
aquatic environment.
The emergence of a cracked actuator housing represents a rare and
unforeseeable event within the operational context. The underlying
factors leading to such damage remain elusive, as it does not stem from
any discernible mistakes or oversights on the part of the operator.
Consequently, the responsibility lies not with human error but rather
with inherent factors that are beyond immediate control.
In summary, due to the highly unpredictable nature of actuator
housing cracking, the responsibility lies not with operator error but with
unanticipated factors. To prevent accidental oil leakage, it is crucial to
diligently observe and detect any signs of cracks before commencing
operations, particularly prior to submerging the ROV into water. Luckily
Fig. 18. Visible crack on a body of the actuator [author’s photos]. ROV operators has daily visual checks of the ROV systems among other
planned works. It ensures timely identification and appropriate mea­
3. Results and discussion sures to address actuator housing cracks or other mechanical damages,
safeguarding the system and minimizing potential environmental
a) Jaw and Tool actuator consequences.

The initial point of contact in any manipulator operation is the jaws. e) Equipment/tools and material used to repair damages on the
It is possible for the operator to misjudge the distance between the manipulator
manipulator and the object, resulting in unintentional contact or
catching the object in between the jaws. In such cases, the operator may Table 5 presented detailed information from the manufacturer,
exert force to position the manipulator, leading to potential damage. Schilling FMC Technologies, regarding the device and materials used to
Consequently, whenever the manipulator jaws sustain damage, it in­ repair damage to the manipulators, along with their catalog numbers.
dicates incorrect control of the manipulator by the operator.

10
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

Table 5
Special tools and materials from Schilling FMC Technologies required
B= 20
to repair damaged manipulators.
where: A – number of failures of a given element for a period of 5 years.
Component Part number

Seal Installation Tool 010–0184 B – number of tested manipulators over 5 years

Seal Re-sizing Tool 010–0185
Sizing tool 010–0203
As an illustrative example, the calculation and graphical represen­
Bearing Installation Tool 010–1457
1 5/8 Crows Foot Wrench 010–0759 tation are provided for damages occurring in the jaws of the manipu­
O-ring Installation Tool 010–0194 lator. Similar calculations are performed for other components of the
Slip ring Re-sizing Tool 010–0220 manipulator in a corresponding manner.
Bearing Pusher 010–0236
e.g. Jaw and Tool actuator.
CV Plug Tool 010–0474
Loctite Primer (Item 23) 002–0291
The number of damages observed in the jaw and tool actuator of the
Loctite 271 50 ml (Item 21) 002–0795 twenty examined T4 manipulators over a five-year period is 16.
DC 55 O-ring Lubricant 002–0359
Aqua Lube 002–0805 16
p=
Petroleum jelly 002–5543 20
Loctite 271 Red 002–0795
O-ring lube (Dow Corning® 55) 002–0359 p= 0.8
Valvoline® Val-plex EP® grease 002–3051
Loctite 609 Green 002–1032 Based on calculations performed in the Geogebra software (Fig. 19),
it can be inferred that the highest probability, rounded to 11.9%, cor­
responds to the quantity of 56. This implies that over a period of 5 years,
3.1. Probability of damage there is a 11.9% probability of encountering damages in 56 jaw and tool
actuators.
Based on the research conducted over a five-year period (Table 3), The outcomes of the probability calculations are presented in detail
the probability of occurrence for individual damages was calculated in Table 6, showcasing the calculated probabilities for each assessed
using the binomial distribution in the Geogebra software. The study was element. This table provides a comprehensive overview of the proba­
conducted in a company that owns and actively operates 37 vessels, with bilities associated with different quantities of damages in all assessed
40 WCROV across a fleet (some of the vessels has two WCROV), all elements of the manipulator. By examining the values presented in
utilizing a manipulator T4 model. Based on authors own experience, it Table 6, valuable insights can be gained regarding the likelihood of
was assumed that each vessel possesses at least one spare T4 manipu­ specific damage occurrences within the given time frame (5 years).
lator. It was further assumed that the number of spare manipulators is The results obtained from the Geogebra software, used a binomial
three-quarters of the total number of manipulators. probability, indicate the most probable quantity of damages, repre­
Consequently, the total number of T4 manipulators employed within senting the combination of the highest probability of damage for a
the company is 70 units (40 multiplied by 3/4). specific component and the maximum number of occurrences. However,
to avoid relying solely on a single precise calculation result, a confidence
3.1.1. Geogebra software interval (2) is subsequently applied, providing a more reliable range of
Geogebra is a widely used software application designed to facilitate probabilities for future damage occurrences.
dynamic mathematical operations across various mathematical disci­ Equation used:
plines such as algebra, statistics, and differential calculus. It also en­ √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
̅ √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
̅⎞
⎛ ( ) ( )
compasses functionalities like calculators, graphing tools, and geometry 2 k
⎜2k + u − u 4k 1 − + u2 2k + u2 + u 4k 1 − k
+ u2 ⎟
tools (GeoGebra). In the context of this study, Geogebra was employed ⎜ n n
⎟
to perform probability calculations regarding the occurrence of damages ⎜
⎜ , ⎟
⎟ (2)
⎝ 2(n + u2 ) 2(n + u2 ) ⎠
on specific elements of the manipulator, utilizing a dataset spanning a
period of five years. Standard settings of the geogebra program from
www.geogebra.org/classic#probability were used.
Where:
parameter n - for all calculations the same. n = 70
parameter p was taken from Table 3.
k – the number of failures of a given element for a period of 5 years
n – number of tested manipulators over 5 years
3.1.2. Calculating probability using geogebra software
α = 0.05 (95%)
To perform calculations in the Geogebra software, it is necessary to
u = 1.95996398
provide parameters such as:
Parameter n and parameter p.
Calculations were performed separately used Microsoft Excel (not
The parameter n will remain constant across all calculations, repre­
shown in the paper).
senting the total number of manipulators used in the company.
Table 7 provides a comprehensive display of confidence intervals
n= 70 pertaining to individual elements, expressed as percentages, coupled
with the specific quantities of probable damages associated with each
n – total number of manipulators used in the company.
element. This table offers valuable insights into the range of uncertainty
The parameter p, however, will vary as it is calculated using formula
surrounding the likelihood of damages occurring in various components
(1), which incorporates the number of failures observed in a specific
of the manipulator.
element of the T4 manipulator over a five-year period.
Additionally, it considers the total number of manipulators tested
during this five-year timeframe, which amounts to 20 units. 3.2. Questionnaire survey
A
p= (1) The questionnaire served to gather additional insights and validate
B
the research conducted by the author. By obtaining responses directly
from the operators who actively utilize the Titan T4 manipulator in their

11
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

Fig. 19. Probability of damage of jaw and tool actuator [Own study].

2. How do you rate T4 according to work with them? Any pros and
Table 6
cons?
Results of probability calculations.
3. What would you change on T4?
Number of damages in total for period of 5 years

Probability The number of damage Upon obtaining responses from the ROV operators, several key ob­
[%] occurrences servations and findings were identified. The analysis of the question­
1. Azimuth 22 3 naire data revealed valuable insights into the performance and
2. Upper Arm 0 0 reliability of the Titan T4 manipulator, thereby serving as a significant
3. Shoulder Actuator 10,4 49 contribution to the overall research objectives.
4. Elbow Actuator 9,5 35
Summarized answers from questionnaires were as follows:
5. Forearm 15,7 7
6. Pitch and Yaw 9,5 31
Actuators 1. Q: From your own experience, what are the most common faults/
7. Wrist Actuator 11,9 14 defects on T4?
8. Jaw and Tool Actuator 11,9 56
9. Compensator 13,3 10
10. Camera 22 3
A: The data collected by the author has been corroborated.
11. Master Controller 10,4 21 The analysis of responses obtained from ROV operators regarding
their own experiences revealed consistent findings that corroborate the
author’s data. The most frequently reported type of damage in the T4
Table 7
manipulator was identified as jaw damage. Notably, respondents high­
Confidence intervals for individual components. lighted that such damage is comparatively less expensive to address and
can be promptly repaired or replaced. This observation underscores the
Confidence intervals for individual components
relative ease and efficiency associated with addressing jaw-related
Confidence intervals Amount of probable faults, contributing to the overall resilience and maintainability of the
[%] damage
T4 manipulator system. By promptly attending to jaw damage, operators
Will get damaged: can swiftly restore the functionality of the manipulator, minimizing
1. Azimuth 0,1–23,6 1–16 operational downtime and ensuring uninterrupted performance during
2. Upper Arm 0,0–16,1 0–11 critical tasks.
3. Shoulder Actuator 48,1–85,5 34–60
4. Elbow Actuator 29,9–70,1 21–49
2. Q: How do you rate T4 according to work with them? Any pros and
5. Forearm 2,8–30,1 2–21
6. Pitch and Yaw 25,8–65,8 18–46 cons?
Actuators
7. Wrist Actuator 8,1–41,6 6–29 A: In response to the question evaluating the performance of the T4
8. Jaw and Tool Actuator 58,4–91,9 41–64
manipulator, all participants provided consistently positive feedback.
9. Compensator 5,2–36,1 4–25
10. Camera 0,1–23,6 1–16
The respondents uniformly expressed their high regard for the T4
11. Master Controller 14,6–51,9 10–36 manipulator, deeming it the premier choice within its designated range
of applications. The overwhelmingly positive assessments affirm the
manipulator’s superior capabilities and effectiveness in executing a
daily operations, it aimed to corroborate the consistency and accuracy of diverse array of tasks. The unanimous consensus regarding its superi­
the reported damages. The questionnaire encompassed various aspects ority serves as a testament to the robust design, functionality, and
related to the manipulator’s performance, potential issues encountered overall quality of the T4 manipulator, further solidifying its reputation
during its operation, and the frequency of damages experienced. as a leading solution in the field of manipulator technology. These
The questions from the questionnaire were as follows: affirmative evaluations underscore the significant role the T4 manipu­
lator plays in enhancing operational efficiency, productivity, and the
1. From your own experience, what are the most common faults/de­ overall success of underwater activities conducted by the operators in
fects on T4? their professional capacities.

12
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

3. Q: What would you change on T4? longevity and proper functioning of the O-rings and other components.
Lastly, but of significant importance, it should be noted that a
A: In response to the question regarding potential improvements or considerable majority of damages are attributed to operator errors,
modifications to the T4 manipulator, the participants offered diverse indicating the critical role of proper operator control in mitigating
suggestions. Among the suggestions put forward, one recurring recom­ damages.
mendation was the utilization of an alternative T4 controller. The pro­ Understanding the vulnerabilities and failure patterns of the Titan 4
posed controller was described as offering enhanced functionality, manipulator’s elements plays a crucial role in optimizing operational
including improved ease of control, a wider range of programmable efficiency, reducing downtime, and enhancing the overall reliability of
functions, and a unique feature of displaying a 3D model representation underwater intervention tasks.
of the slave arm’s movements in space, effectively illustrating stress
points. Funding
It is worth noting that the suggested controller is not an official
product offered by Schilling, the manufacturer of the T4 manipulator. This research funded by the Ministerstwo Edukacji i Nauki, Polska,
Nonetheless, the respondents expressed a desire for a controller that grant number 1/S/KE/23.
combines advanced control capabilities with visual representations that
aid in better understanding the manipulator’s movements and stress CRediT authorship contribution statement
distribution. The proposed enhancements align with the operators’ as­
pirations for increased control precision, enhanced programmability, Dalibor Ivanega: Conceptualization, Methodology, Software, Vali­
and improved situational awareness during complex underwater dation, Formal analysis, Investigation, Resources, Data curation, Visu­
operations. alization, Project administration, Writing – original draft, Writing –
review & editing. Marcin Szczepanek: Conceptualization, Methodol­
4. Conclusion ogy, Validation, Formal analysis, Investigation, Resources, Writing –
original draft, Writing – review & editing, Supervision, Project admin­
The aim of this study was to assess the operational characteristics of istration, Funding acquisition.
the Schilling Titan 4 manipulator by estimating the most frequently
occurring damages on its individual components. The relevant elements Declaration of competing interest
include the azimuth, shoulder, upper arm, elbow, forearm, pitch & yaw,
wrist, jaw, compensator, camera, and master controller. Based on the The authors declare that they have no known competing financial
specific damages observed by the author over a five-year period on the interests or personal relationships that could have appeared to influence
tested twenty manipulators, the jaws (jaw) were identified as the most the work reported in this paper.
commonly damaged component, with 16 reported damages. Following
that, the shoulder actuator, responsible for lifting the manipulator arm, Data availability
exhibited 14 damages, while the elbow actuator displayed 10 damages,
making them the most vulnerable elements in the Titan 4 manipulator. No data was used for the research described in the article.
One of the objectives of this study was also to explore potential
preventive measures against such damages in the future. Upon a thor­ References
ough analysis of the observed damages, it can be concluded that, for
example, the design of the jaws prioritizes their susceptibility to damage Aldhaheri, S., De Masi, G., Pairet, È., Ardón, P., 2022. Underwater Robot Manipulation:
Advances, Challenges and Prospective Ventures. IEEE, pp. 1–7.
over other more complex, expensive, and time-consuming elements. The Antonelli, G., 2014. Underwater Robots, vol. 96. Springer Tracts in Advanced Robotics.
jaws are specifically designed to facilitate swift repairs or replacements, Bird, B., Griffiths, A., Martin, H., Codres, E., Jones, J., Stancu, A., Lennox, B., Watson, S.,
eliminating the need to disconnect hydraulic lines, drain the system’s Poteau, X., 2018. A robot to monitor nuclear facilities: using autonomous radiation-
monitoring assistance to reduce risk and cost. IEEE Robot. Autom. Mag. 26, 35–43.
oil, or engage in further laborious steps. Consequently, the jaws can be Bogue, R., 2015. Underwater robots: a review of Technologies and applications. Ind.
replaced in a matter of minutes, allowing the ROV to resume its oper­ Robot: Int. J. 42, 186–191.
ations promptly, without experiencing the significant downtime typi­ Capocci, R., Dooly, G., Omerdić, E., Coleman, J., Newe, T., Toal, D., 2017. Inspection-
class remotely operated vehicles—a review. JMSE 5, 13. https://doi.org/10.3390/
cally associated with repairs. Positioned at the very end of the
jmse5010013.
manipulator, the jaws constitute the initial point of contact during every Christ, R.D., Wernli, R.L., 2014. The ROV Manual: A User Guide for Remotely Operated
operation, underscoring the critical role of operator experience in Vehicles, second ed. Butterworth-Heinemann is an imprint of Elsevier, Amsterdam.
responsibly manoeuvring the manipulator to avoid damages. 978-0-08-098288-5.
Cochran, J.K. (Ed.), 2019. Encyclopedia of Ocean Sciences, third ed. Academic Press is an
The shoulder actuator, responsible for lifting the manipulator arm, is imprint of Elsevier, London, United Kingdom ; Cambridge, MA, United States. 978-0-
often subject to damages when operators attempt to lift objects 12-813081-0.
exceeding the manipulator’s weight limit or due to accidental collisions Coleman, D.F., Ballard, R.D., Gregory, T., 2003. Marine archaeological exploration of the
black sea. IEEE 3, 1287–1291.
with underwater structures, entanglements, and subsequent extraction Dalhatu, A.A., Sa’ad, A.M., Cabral de Azevedo, R., De Tomi, G., 2023. Remotely operated
attempts. Repairing or replacing the hydraulic actuator is a more time- vehicle taxonomy and emerging methods of inspection, maintenance, and repair
consuming process compared to replacing the jaws. operations: an overview and outlook. J. Offshore Mech. Arctic Eng. 145, 020801.
Davey, V.S., Forli, O., Raine, G., Whillock, R., 1999. Non-Destructive Examination of
Similarly, the elbow actuator, like the other actuators within the Underwater Welded Structures, vol. 1372. Woodhead Publishing, 1-85573-427-3.
manipulator’s joints, is a highly complex component that demands a Fletcher, B., 2000. Worldwide Undersea MCM Vehicle Technologies, vol. 11. Space and
lengthy repair process. Typically, the entire manipulator is replaced Naval Warfare System Center, San Diego, CA. Rep. MS.
GeoGebra. Available online: https://www.geogebra.org/about. (Accessed 26 July 2023).
with a spare unit, and the damaged elbow actuator (or any other faulty Guo, M., Tian, Y., 2022. Review on structural design of underwater manipulator. SPIEL
joint) is overhauled on the defective manipulator. 12309, 310–315.
Upon analysing all the reported damages, a common factor emerges, Harden, T.A., Lloyd, J.A., Turner, C.J., 2009. Robotics for Nuclear Material Handling at
LANL: Capabilities and Needs.
namely leaks or sealing failures. The Titan 4 manipulator incorporates
Huang, H., Tang, Q., Li, J., Zhang, W., Bao, X., Zhu, H., Wang, G., 2020. A review on
numerous O-rings of varying types and dimensions, and it is crucial to underwater autonomous environmental perception and target grasp, the challenge of
exercise caution to avoid mistakenly interchanging similar O-rings with robotic organism capture. Ocean. Eng. 195, 106644.
slight numerical variations. Kennish, M.J., 2019. Practical Handbook of Marine Science. crc press, 1-351-65410-1.
Mazzeo, A., Aguzzi, J., Calisti, M., Canese, S., Vecchi, F., Stefanni, S., Controzzi, M.,
Adhering to the manufacturer’s recommendations, regular in­ 2022. Marine robotics for deep-sea specimen collection: a systematic review of
spections of the manipulator should be conducted to ensure the underwater grippers. Sensors 22, 648.

13
D. Ivanega and M. Szczepanek Ocean Engineering 291 (2024) 116282

McLean, D.L., Parsons, M.J., Gates, A.R., Benfield, M.C., Bond, T., Booth, D.J., Bunce, M., Simetti, E., Wanderlingh, F., Torelli, S., Bibuli, M., Odetti, A., Bruzzone, G., Rizzini, D.L.,
Fowler, A.M., Harvey, E.S., Macreadie, P.I., 2020. Enhancing the scientific value of Aleotti, J., Palli, G., Moriello, L., 2017. Autonomous underwater intervention:
industry remotely operated vehicles (ROVs) in our oceans. Front. Mar. Sci. 7, 220. experimental results of the MARIS project. IEEE J. Ocean. Eng. 43, 620–639.
Nichols, C.R., Williams, R.G., 2009. Encyclopedia of Marine Science. Infobase Publishing, Sivčev, S., Coleman, J., Omerdić, E., Dooly, G., Toal, D., 2018a. Underwater
1-4381-1881-3. manipulators: a review. Ocean. Eng. 163, 431–450. https://doi.org/10.1016/j.
O’Donnell, W.M., 1995. Automated Robotics System for Nuclear Waste Handling oceaneng.2018.06.018.
(Hardware and Software). University of Nevada, Las Vegas, 9798209341680. Sivčev, S., Rossi, M., Coleman, J., Omerdić, E., Dooly, G., Toal, D., 2018b. Collision
Petillot, Y.R., Antonelli, G., Casalino, G., Ferreira, F., 2019. Underwater robots: from detection for underwater ROV manipulator systems. Sensors 18, 1117.
remotely operated vehicles to intervention-autonomous underwater vehicles. IEEE Teague, J., Allen, M.J., Scott, T.B., 2018. The potential of low-cost ROV for use in deep-
Robot. Autom. Mag. 26, 94–101. sea mineral, ore prospecting and monitoring. Ocean. Eng. 147, 333–339.
Ribas, D., Palomeras, N., Ridao, P., Carreras, M., Mallios, A., 2011. Girona 500 AUV: TechnipFMC. Schilling Robotics TITAN 4 Manipulator.
from survey to intervention. IEEE ASME Trans. Mechatron. 17, 46–53. Thompson, O.; Knight, A. An Overview of the Work Class ROV Market Available online:
Sánchez, P.J.B., Papaelias, M., Márquez, F.P.G., 2020. Autonomous underwater vehicles: https://www.oedigital.com/news/489411-an-overview-of-the-work-class-rov-mar
instrumentation and measurements. IEEE Instrum. Meas. Mag. 23, 105–114. ket (accessed on 18 July 2023).

14