IAES International Journal of Robotics and Automation (IJRA)

Vol. 14, No. 1, March 2025, pp. 1~10

ISSN: 2722-2586, DOI: 10.11591/ijra.v14i1.pp1-10  1

Position and orientation analysis of Jupiter robot arm for

navigation stability

Omar Shalash1,2, Adham Sakr1, Yasser Salem1, Ahmed Abdelhadi1, HossamEldin Elsayed1,2,
Ahmed El-Shaer1
1
College of Artificial Intelligence, Arab Academy for Science, Technology and Maritime Transport, Alamein, Egypt
2
Research and Innovation Center, Arab Academy for Science, Technology and Maritime Transport, Alamein, Egypt

Article Info ABSTRACT

Article history: Jupiter robot has made a great impact in the educational field with its
support for autonomous navigation, visual perception, and many other
Received Jul 19, 2024 features from its artificial intelligence platform's learning box. This study
Revised Sep 7, 2024 undertakes a kinematic model design of Jupiter's arm to aid the robot's
Accepted Sep 17, 2024 motion stability. This process involved the determination of a homogeneous
transformation matrix, followed by the determination of orientation,
position, and Euler angles. Ultimately, the homogeneous transformation
Keywords: matrix was successfully derived, and the simplification of direct kinematic
matrices was achieved. Consequently, the kinematic analysis for Jupiter's
Analytical solutions arm was established using the position Denavit–Hartenberg method,
Denavit–Hartenberg orientation, and Euler angles, proving to be valuable in the context of this
Inverse kinematics research.
Jupiter robot
Kinematic analysis This is an open access article under the CC BY-SA license.

Corresponding Author:
Omar Shalash
College of Artificial Intelligence, Arab Academy for Science, Technology and Maritime Transport
Alamein, Egypt
Email: [email protected]

1. INTRODUCTION
Robotics, the interdisciplinary field at the intersection of mechanical engineering, computer science,
artificial intelligence, and electronics, has rapidly transformed the way we live and work. In the quest for
automating tasks and improving abilities, robotics has become a technology in the 21st century. Originating
from the realm of science fiction and finding its way into industries such, as manufacturing and healthcare
artificial intelligence has become widely prevalent [1]–[13]. The study of robotics has become a dynamic and
critical area of research, offering insights into designs, developments, and deployments of intelligent
machines.
The field of robotics has made strides thanks to advancements in modeling techniques. It is crucial
to comprehend the positioning and orientations of robots such, as the Jupiter robot in order to effectively
control and interact with them in environments. Two key components, forward kinematics and inverse
kinematics play a role in this pursuit. Significant research has been devoted to finding solutions for
determining the kinematics of humanoid robots. In the past researchers primarily utilized Denavit–Hartenberg
(DH) parameters, which offered a standardized approach, for representing the transformations, between
neighboring links [14]. In this approach, we assign parameters to each joint allowing us to calculate the
positions and orientations of the end effector [15]–[21]. Another used technique in robotics is the
homogeneous transformation matrices. They provide a framework, for expressing the position and orientation
transformations of robotic systems [22].

Journal homepage: http://ijra.iaescore.com

2  ISSN: 2722-2586

These matrices make computations more efficient, by containing all the transformation information
within a matrix [23]–[28]. The formula known as the product of exponentials which was introduced by
Murray et al. [29] is a representation that is derived from Lie algebra. It provides a compact and elegant way
to express the end-effector pose as a function of joint variables, allowing for efficient and accurate kinematic
computations [30]–[33]. Finding equations, for particular robot configurations is especially valuable when
dealing with robots that have clearly defined and relatively uncomplicated shapes [34]. They offer explicit
equations for determining end-effector poses [35]–[37]. The flexible method of inverse kinematics and
trigonometric methods utilizes inverse trigonometric functions to calculate the angles of joints. In particular,
this technique is very straightforward, as the corresponding joints’ orientations can easily determine the
angles needed to achieve certain target locations of an end-effector [37]–[39]. A representation based on the
idea of Geometric interpretation of kinematics clarifies to us how the motions of joints are related with the
motion of the end effector. It is a way of conceptualizing robot motion and understanding how everything fits
together [40], [41]. Currently, these solutions have been expanded into dealing with kinematically complex
structures as well as redundant systems. With this development, analysis solutions have taken on a new
dimension; more flexible and complex robotics systems.
Determining the angles of joints to find the suitable position of an end effector, known as
kinematics, has been a crucial field of study. Researchers have utilized methods, like techniques and
optimization algorithms, to address this complex problem. These methods have shown their effectiveness, in
real-time control and planning movements [42]. Iterative Jacobian-based methods are a prevalent approach in
solving inverse kinematics [43]. They use the matrix, which helps determine how changes, in velocities,
affect the velocities of the end effector [44]. Understanding the relationship between angles and the resulting
changes in the position and orientation of the end effector is essential. This matrix serves as a valuable tool,
for that purpose [45]. By making modifications these techniques gradually move closer, to the intended
position skillfully maneuvering through the range of solutions and steadily moving towards the objective
[45], [46]. They perform well in changing environments and situations that demand immediate adaptations.
However, their efficiency can be affected by factors such, as the iteration method used the accuracy of
calculating the Jacobian, and the occurrence of singularities [44]. Reliable convergence depends on careful
parameter tuning. However, the choice between the Iterative Jacobian-based methods and other techniques
depends on various considerations which include robot structures, computational resources, and operating
environment. In making the right choice about various methods regarding inverse kinematics, it is important
to understand each method’s strong points as well as its weak points [47]. The other important approach is
that the CCD algorithm has been a significant approach for inverse kinematics. Through an iterative
adjustment of each separate joint’s angle, CCD is developed specifically for the articulated structures
possessed by several degrees of freedom robots [48]. The operation begins on its base and improves
successive angles on the kinetic chain till it gets to the position it desires for the end effector. This technique
takes advantage of the intrinsic organization of many robotic systems and is therefore highly useful for
solving complicated articulations like those characterizing humanoid robots as well as multi-limbed robotic
arms [49]. Nevertheless, it’s important to acknowledge that while CCD excels in many scenarios, it may face
challenges converging to a solution in cases involving singularities or highly constrained environments [50].
Additionally, the order of joint adjustments can impact the outcome, warranting thoughtful consideration in
its application. Despite these considerations, CCD remains a valuable and pragmatic tool for addressing
complex inverse kinematics challenges in robotics as reported in Table 1 [51]. Gradient-based optimization
approaches inverse kinematics as an optimization challenge, employing techniques like gradient descent and
Newton’s method to minimize an objective function, hence converging towards optimal joint angles [52].
The process whereby joint angles are adjusted incrementally moving the system along the path of decreasing
the value of the objective function until reaching a local minimum that coincides with the target end-effector
positioning is referred to as gradient decent. It has been observed that some complex and non-linear objective
functions, could converge faster than first-order derivatives using Newton’s methods [53]. It may be strong
but you need to consider the choice of an optimization algorithm, a convergence criterion, and the objective
function definition. Since singularities and constrained environments require special handling [49].
Knowledge of such subtleties is crucial in the effective utilization of addressing the inverse kinematics
problem in robotics. Damped least squares (DLS) is an approach with both analytical and numerical
ingredients applied within inverse kinematics [54]. It optimizes its weighted least square to solve the problem
of minimizing this difference between the end-effector poses. Therefore, one should include this damping
factor that will provide enough stability that make the solution correct under such instabilities or arising
inconsistencies [50]. The advantage of Damped Least Squares is inaccurate end-effector poses for the
systems that use it. It works out to be a good choice when it comes to robotic applications such as
manipulator arms and humanoid robots [55]. Researchers have delved into diverse optimization-based
techniques, such as genetic algorithms, simulated annealing, and particle swarm optimization (PSO), as

IAES Int J Rob & Autom, Vol. 14, No. 1, March 2025: 1-10
IAES Int J Rob & Autom ISSN: 2722-2586  3

alternative avenues for tackling inverse kinematics challenges [54]. The use of these methods brings about
another paradigm because they apply to nature-based search strategies. Genetic algorithms use principles of
natural selection and genetics for exploring the solution space and hence are very effective in solving
difficult, multidimensional optimization problems [56]. These methods offer diverse strategies for tackling
the complex problem of inverse kinematics, each with its own strengths and considerations. Their application
depends on factors such as the robot’s stability kinematic structure, computational resources, and specific
task requirements.
This research contributes by developing a kinematic model for the Jupiter robot, enhancing its arm
functionality in several key ways. First, the model allows the robot's arm to extend fully, providing an
advantage in competitive settings by enabling faster object-grabbing. Second, the kinematic model improves
the stability of the robot's movements, ensuring smooth and controlled operation. These features allow the
Jupiter robot to operate at maximum speed without risking tipping over, while swiftly reaching and grasping
target objects.

2. JUPITER BACKGROUND
The Jupiter robot as seen in Figure 1, created by Lattel Robotics, is a versatile home assistant
weighing 10.3 kg and boasting external dimensions of 352×352×920 mm, with a ground clearance of 15 mm.
Lattel Robotics, a company dedicated to promoting AI-focused robotics education, is headquartered in
Singapore and Malaysia. The heart of the Jupiter robot is its onboard computer, equipped with an Intel Core
i5-10210U processor running at 1.6 GHz, 8 GB of RAM, and a 120 GB SSD for internal storage. It also
features a Wi-Fi remote controller with a swift 300 Mbps transmission rate, enabling seamless interaction.
Speech interaction capabilities are incorporated, offering a frequency response between 50 Hz and 16 kHz for
clear communication. For mobility, the robot utilizes the mobile base Kobuki unit, allowing for a maximum
payload of 5 kg, a top speed of 0.5 m/s, and a rapid rotation speed of 160 degrees per second. Powering this
remarkable machine are standard 4400mAh Li-Ion batteries, with the option for extended battery life using an
additional 4400mAh Li-Ion unit. The robot is equipped with an array of sensors, including 25,700 CPR
encoders, a rate gyroscope with a factory calibration of 110 deg/s, and auxiliary sensors (3x forward bump,
3x Cliff, 2x wheel drop). It also features a 3D stereo camera with a resolution of 640px x 480px, recording at
30 fps, and a Slamtec A2m8-R4 RP LiDAR system. With all these features, the Jupiter robot is equipped with
various support for autonomous navigation, visual perception, speech interaction, mobile manipulation and
AI, machine learning, and cloud computing, hence proven to be essential in the educational field, which
enabled the robot to be the core element in RoboCup@Home Education contest [57].

Figure 1. Jupiter robot

Position and orientation analysis of Jupiter robot arm for navigation stability (Omar Shalash)
4  ISSN: 2722-2586

Jupiter’s arm system as seen in Figure 1 is comprised of 5 servos, including the end effector servo,
along with 4 links. Its external dimensions (L×W×H) are 32×50×40 mm, and it offers an impressive accuracy
of 0.29 degrees [57]. This robotic wonder seamlessly integrates a blend of features, including robot
intelligence, natural interaction, computer vision, mobile platform capabilities, and object manipulation. The
robot’s ARM is a 4DOF (four degrees of freedom) system. It comprises four joints, each with its unique
rotation: Joint 1: Yaw rotation, Joint 2: Pitch rotation, Joint 3: Pitch rotation, Joint 4: Pitch rotation, These
four joints collectively endow the robot’s effector with four degrees of freedom, enabling motion in the x, y,
and z axes, as well as pitch and yaw rotation.

3. METHODOLOGY
The transformation matrix for joint ‘i‘ with respect to an adjacent joint ‘i‘ in three-dimensional space
is in (1).

𝑐𝑜𝑠(𝑖 ) −𝑠𝑖𝑛(𝑖 ) 0 𝑖−1

𝑐𝑜𝑠(𝑖−1 )𝑠𝑖𝑛(𝑖 ) 𝑐𝑜𝑠(𝑖−1 )𝑐𝑜𝑠(𝑖 ) −𝑠𝑖𝑛(𝑖−1 ) −𝑑 𝑠𝑖𝑛(𝑖−1 )
𝑇𝑖 = [ ] (1)
𝑠𝑖𝑛(𝑖−1 )𝑠𝑖𝑛(𝑖 ) 𝑠𝑖𝑛(𝑖−1 )𝑐𝑜𝑠(𝑖 ) 𝑐𝑜𝑠(𝑖−1 ) 𝑑 𝑐𝑜𝑠(𝑖−1 )
0 0 0 1

The transformation from joint ‘i‘ to joint ‘j‘, where ‘i‘ ranges from 1 to ‘N‘, and ‘j‘ ranges from ‘i+1‘ to ‘N‘,
is given by (2).

𝑗 𝑗
𝑇𝑖 = 𝑇𝑖𝑖+1 ⋅ 𝑇𝑖+1
𝑖+2
⋅ … ⋅ 𝑇𝑗−1 (2)

For the case of a 5-joint robotic arm as observed in Figure 2, the transformation matrices are filled for each
joint (2), and the multiplication is processed sequentially to obtain the transformation from the base to the
end-effector frame (3).
4 4 4 4
0𝑅11 0𝑅12 0𝑅13 0𝑋
4 4 4 4
0 0𝑅21 0𝑅22 0𝑅23 0𝑌
4𝑇 = 4 4 4 4
(3)
0𝑅31 0𝑅32 0𝑅33 0𝑍
[ 0 0 0 1]

In order to prepare the DH model for Jupiter’s arm, see Figure 2 for illustration, the model’s parameters
(the link length (ai), link offset (di), rotation angle (θi), and link twist (αi)) are calculated and presented in
Table 1.
Solving the inverse kinematic problem depends on each robot’s design. Iterative numerical methods
may eventually lead to a solution, but they are prone to encountering singularities, resulting in potential
failures even when a valid solution is possible. Furthermore, their performance is generally sub-optimal. On
the other side, analytical solutions are swift and highly precise, yet uncovering them necessitates a substantial
amount of effort.

Figure 2. Five-joint robotic arm

IAES Int J Rob & Autom, Vol. 14, No. 1, March 2025: 1-10
IAES Int J Rob & Autom ISSN: 2722-2586  5

Table 1. DH-model parameters

i αi−1 ai−1 di θi
1 0 0 a1 θ1
2 −90° 0 0 θ2
3 0 L2 0 θ3
4 0 L3 0 θ4
5 0 L4 0 0

3.1. Geometric solution

The geometric solution for inverse kinematics was chosen according to two factors. First, it is an
Efficient solution for simple structures with closed-form solutions, thus greatly reducing many calculations
and increasing efficiency. Second, the geometric solution is adaptable, allowing more modifications to the
arm without requiring any changes to the inverse kinematics model, as long as the parameters and
coefficients stay the same. In comparison to other methods, such as the algebraic track, the geometric solution
is valued for its efficiency in less complex structures. While the algebraic method boasts greater versatility
and can handle a broader spectrum of robotic systems, it may grapple with challenges, particularly in
scenarios that involve highly nonlinear systems [58]. Moreover, when compared to numerical solutions such
as neuro-fuzzy models, aside from being known for their adaptability, they introduce complexities related to
computational costs, interpretability, and data requirements. The decision to choose one of these models
depends on the needs of the robotic system. Necessitating a careful balance between adaptability,
computational requirements, and interpretability [59] and comparison with new techniques, such as the use of
PSO. combined with POSIX threads for inverse kinematics. While this innovative strategy holds promise by
facilitating parallelism and potential speedup, it presents challenges related to thread synchronization, load
balancing, and system complexity. Rigorous testing and validation under various conditions is essential to
accurately evaluate its performance [59]. After carefully considering these factors, the geometric solution
approach emerged as the preferred choice in this study. Its efficiency in handling simple structures and
closed form solutions, and the adaptability of the geometric solution allow for dynamic changes in link
lengths or the end-effector while maintaining the same configuration. This feature is particularly beneficial
in scenarios where alterations to the robot’s structure are required. It is the most fit solution for Jupiter’s
robotic system and its unique requirements. The top view of the robotic arm, the value of r1, using the
Pythagorean theorem in (5).
Given that yo is the position of the end effector along the y-axis and xo is the position of the end
effector along the x-axis.

𝑦𝑜
θ4 = tan−1 ( ) (4)
𝑥𝑜

θ1 represents the rotational joint of the arm (base joint). Given that, z0 is the distance the end effector along
the z-axis, and a0 is the base link length. The side view of the robot’s arm:

𝑟1 = √𝑥02 + 𝑦𝑜2 ; 𝑟2 = 𝑧0 − 𝑎0 ; 𝑟3 = √𝑟12 + 𝑟22 (5)

By applying the cosine rule (6) on the r3 link

(𝑟32 = 𝑙12 + 𝑙22 − 2 ⋅ 𝑙1 ⋅ 𝑙2 ⋅ cos α) (6)

The angle α can be denoted by (7)/

𝑙12 + 𝑙22 − 𝑟32

cos α = (7)
2 ⋅ 𝑙1 ⋅ 𝑙2

θ2 and α are supplementary and can be obtained by (8).

𝑟12 + 𝑟22 − (𝑙12 + 𝑙22 )

θ2 = (π − α); cos θ2 = − cos α ; θ2 = cos −1 ( ) (8)
2 ⋅ 𝑙1 ⋅ 𝑙2

Then, applying (SOHCAHTOA) trigonometric functions (9) to deduce the angles equation.

Position and orientation analysis of Jupiter robot arm for navigation stability (Omar Shalash)
6  ISSN: 2722-2586

𝑙2 sin θ2 𝑟2 𝑙2 sin θ2 𝑟2
θ1 = γ − β; tan β = ; tan γ = ; β = tan−1 ( ) ; γ = tan−1 ( ) (9)
𝑙1 + 𝑙2 cos θ2 𝑟1 𝑙1 + 𝑙2 cos θ2 𝑟1

This represents the result of substituting previous equations to result in (10).

𝑟2 𝑙2 sin 𝜃2
𝜃1 = tan−1 ( ) − tan−1 ( ) (10)
𝑟1 𝑙1 + 𝑙2 cos 𝜃2

Finally, we calculate the angles of movement in the equations below where θbase represents the base rotation
angle, and φ is the Euler angle.

𝑌
𝜃benc = tan−1 ( ) ; 𝜑 = 𝜃1 + 𝜃2 + 𝜃3 ; 𝜃3 = 𝜑 − (𝜃1 + 𝜃2 ) (11)
𝑋

4. RESULTS
A set of tools and libraries have been used to implement and test the robot’s arm in a simulation.
First, a visualization of the robot’s robotic arm using the ”Visualize Kinematics” library on Python as seen in
Figure. In this library in Python, the kinematics model is implemented and tested. The robot’s end-
effector is given a location and orientation, these coordinates are compared with the origin coordinate frame
to prove accuracy.

Figure 3. Kinematics visualization

In the proposed study, we employed a versatile robot equipped with an advanced arm and a multi-
Astra depth camera. The experiments were conducted in diverse environments, such as kitchen and living
room settings, to assess the robot’s ability to autonomously navigate, detect different target objects with
different dimensions and grasping points and center of mass, and perform precise arm manipulations, for the
arm’s position see Figure 3. The robot’s end effector was designed for adaptability, allowing for the grasping
of various objects and their delivery to the user.
Notably, the robot’s end effector is a modular component that can be changed or modified to
perform various tasks. The adaptation also involves tweaking of the link lengths which is possible but does
not pose any issue for inverse kinematics. With respect to this feature, the robot becomes more flexible,
performing diverse operations in various settings with high accuracy and performance.

4.1. Final task

Sensors coupled with algorithms enabled the robot to move precisely to the predefined waypoint. At
this point, the robot started 3D localization of the ball on the spot by multiple astra-depth cameras with
respective x, y, and z coordinates. Subsequently, the robot seamlessly transitioned to arm manipulation,
employing inverse kinematics to calculate joint angles for an accurate grasp and relocation of the ball.

IAES Int J Rob & Autom, Vol. 14, No. 1, March 2025: 1-10
IAES Int J Rob & Autom ISSN: 2722-2586  7

4.2. Validation
Validating the accuracy and reliability of the proposed inverse kinematics model is crucial to ensure
its applicability and effectiveness in practical scenarios. To initiate the validation, an experimental setup was
designed by manually positioning the end effector to a predefined point in space and measuring the join
angles θ1,θ2 and θ4, then cross-validating the measured values with the proposed model.
The end effector was placed at point (10,0,14) in space which corresponds to x, y, z coordinates
respectively, then the joint angles were measured θ1 = 0°, θ2 = 90° and θ4 = 0° and given the link lengths of the
robot arm l1 = 10 cm, l2 = 10 cm and a0 = 4 cm. we substitute in (4), (8), (9) to obtain that the result of the
joint angles θ1 = 0°, θ2 = 90° and θ4 = 0°, see Figure 1.

5. CONCLUSION
In order to develop a robust kinematic model for Jupiter’s robotic arm, we presented an exact
forward and inverse kinematics solution, encompassing a step-by-step derivation of a homogeneous
transformation matrix, determination of orientation, position, and Euler angles, culminated in a streamlined
direct kinematic solution, followed by the simplification of direct kinematic matrices. This contribution not
only improves the robot’s motion control but also aligns with its educational objectives by facilitating
autonomous navigation and enhancing learning experiences. While this study presents a solid foundation,
future work could apply the kinematic model to real-life scenarios and adaptive learning mechanisms to
further advance Jupiter’s educational contributions.

REFERENCES
[1] M. S. Kaiser, S. Al Mamun, M. Mahmud, and M. H. Tania, “Healthcare robots to combat COVID-19,” in Lecture Notes on Data
Engineering and Communications Technologies, vol. 60, Springer Singapore, 2021, pp. 83–97. doi: 10.1007/978-981-15-9682-
7_10.
[2] J. Holland et al., “Service robots in the healthcare sector,” Robotics, vol. 10, no. 1, Mar. 2021, doi: 10.3390/robotics10010047.
[3] I. M. Gaber, O. Shalash, and M. S. Hamad, “Optimized inter-turn short circuit fault diagnosis for induction motors using neural
networks with LeLeRU,” Feb. 2023. doi: 10.1109/CPERE56564.2023.10119618.
[4] A. Khaled, O. Shalash, and O. Ismaeil, “Multiple objects detection and localization using data fusion,” Dec. 2023. doi:
10.1109/ICARCE59252.2024.10492609.
[5] O. Shalash, “Design and development of autonomous robotic machine for knee arthroplasty,” Ph.D. dissertation, Department of
Biomedical Engineering, University of Strathclyde, 2018.
[6] S. K. Kavoussi, K. M. Kavoussi, and D. I. Lebovic, “Robotic-assisted tubal anastomosis with one-stitch technique,” Journal of
Robotic Surgery, vol. 8, no. 2, pp. 133–136, Dec. 2014, doi: 10.1007/s11701-013-0442-z.
[7] O. Shalash and P. Rowe, “Computer-assisted robotic system for autonomous unicompartmental knee arthroplasty,” Alexandria
Engineering Journal, vol. 70, pp. 441–451, May 2023, doi: 10.1016/j.aej.2023.03.005.
[8] E. Khatab, A. Onsy, M. Varley, and A. Abouelfarag, “A lightweight network for real-time rain streaks and rain accumulation
removal from single images captured by AVs,” Applied Sciences (Switzerland), vol. 13, no. 1, Dec. 2023, doi:
10.3390/app13010219.
[9] M. Roche, “The MAKO robotic-arm knee arthroplasty system,” Archives of Orthopaedic and Trauma Surgery, vol. 141, no. 12,
pp. 2043–2047, Nov. 2021, doi: 10.1007/s00402-021-04208-0.
[10] B. Yan, P. Fan, X. Lei, Z. Liu, and F. Yang, “A real-time apple targets detection method for picking robot based on improved
YOLOv5,” Remote Sensing, vol. 13, no. 9, Apr. 2021, doi: 10.3390/rs13091619.
[11] A. Abouelfarag, M. A. Elshenawy, and E. A. Khattab, “Accelerating sobel edge detection using compressor cells over FPGAs,” in
Smart Technology Applications in Business Environments, IGI Global, 2017, pp. 1–21. doi: 10.4018/978-1-5225-2492-2.ch001.
[12] H. Said et al., “Forearm intravenous detection and localization for autonomous vein injection using contrast-limited adaptive
histogram equalization algorithm,” Applied Sciences (Switzerland), vol. 14, no. 16, Aug. 2024, doi: 10.3390/app14167115.
[13] M. Elkholy, O. Shalash, M. S. Hamad, and M. S. Saraya, “Empowering the grid: a comprehensive review of artificial intelligence
techniques in smart grids,” in 2024 International Telecommunications Conference, ITC-Egypt 2024, Jul. 2024, vol. 15, pp. 513–
518. doi: 10.1109/ITC-Egypt61547.2024.10620543.
[14] J. Denavit and R. S. Hartenberg, “A kinematic notation for lower-pair mechanisms based on matrices,” Journal of Applied
Mechanics, vol. 22, no. 2, pp. 215–221, Jun. 1955, doi: 10.1115/1.4011045.
[15] C. Klug, D. Schmalstieg, T. Gloor, and C. Arth, “A complete workflow for automatic forward kinematics model extraction of
robotic total stations using the Denavit-Hartenberg convention,” Journal of Intelligent and Robotic Systems: Theory and
Applications, vol. 95, no. 2, pp. 311–329, Sep. 2019, doi: 10.1007/s10846-018-0931-4.
[16] M. Flanders and R. C. Kavanagh, “Build-a-robot: Using virtual reality to visualize the Denavit-Hartenberg parameters,”
Computer Applications in Engineering Education, vol. 23, no. 6, pp. 846–853, Apr. 2015, doi: 10.1002/cae.21656.
[17] A. C. Reddy, “Difference between Denavit - Hartenberg (D-H) classical and modified conventions for forward kinematics of
robots with case study,” in International Conference on Advanced Materials and manufacturing Technologies (AMMT), 2014, pp.
267–286.
[18] P. I. Corke, “A simple and systematic approach to assigning Denavit-Hartenberg parameters,” IEEE Transactions on Robotics,
vol. 23, no. 3, pp. 590–594, Jun. 2007, doi: 10.1109/TRO.2007.896765.
[19] S. Shim, S. Lee, S. Joo, and J. Seo, “Denavit-Hartenberg notation-based kinematic constraint equations for forward kinematics of
the 3-6 Stewart platform,” Journal of Mechanisms and Robotics, vol. 14, no. 5, Mar. 2022, doi: 10.1115/1.4053822.
[20] R. Issa et al., “A data-driven digital twin of electric vehicle Li-Ion battery state-of-charge estimation enabled by driving behavior
application programming interfaces,” Batteries, vol. 9, no. 10, Oct. 2023, doi: 10.3390/batteries9100521.
[21] S. Karmakar and C. J. Turner, “Forward kinematics solution for a general Stewart platform through iteration-based simulation,”
International Journal of Advanced Manufacturing Technology, vol. 126, no. 1–2, pp. 813–825, Feb. 2023, doi: 10.1007/s00170-

Position and orientation analysis of Jupiter robot arm for navigation stability (Omar Shalash)
8  ISSN: 2722-2586

023-11130-9.
[22] S. Kucuk and Z. Bingul, “Robot kinematics: Forward and inverse kinematics,” in Industrial Robotics: Theory, Modelling and
Control, Pro Literatur Verlag, Germany / ARS, Austria, 2006. doi: 10.5772/5015.
[23] O. Martínez and R. Campa, “Comparing methods using homogeneous transformation matrices for kinematics modeling of robot
manipulators,” in Mechanisms and Machine Science, vol. 94, Springer International Publishing, 2021, pp. 110–118. doi:
10.1007/978-3-030-60372-4_13.
[24] T. P. Singh, P. Suresh, and S. Chandan, “Forward and inverse kinematic analysis of robotic manipulators,” International Research
Journal of Engineering and Technology(IRJET), vol. 4, no. 2, pp. 1459–1469, 2017.
[25] J. R. Ortiz-Zacarias, Y. S. Valenzuela-Lino, J. Asto-Evangelista, and D. Huamanchahua, “Kinematic position and orientation
analysis of a 4 DoF orthosis for knee and ankle rehabilitation,” in 2021 6th International Conference on Robotics and Automation
Engineering, ICRAE 2021, Nov. 2021, vol. 18, pp. 141–146. doi: 10.1109/ICRAE53653.2021.9657817.
[26] G. Dudek and M. Jenkin, Computational principles of mobile robotics. Cambridge University Press, 2010. doi:
10.1017/cbo9780511780929.
[27] P. Curtis, C. S. Yang, and P. Payeur, “An integrated robotic multi-modal range sensing system,” in Conference Record - IEEE
Instrumentation and Measurement Technology Conference, 2005, vol. 3, pp. 1991–1996. doi: 10.1109/imtc.2005.1604520.
[28] M. Yasser, O. Shalash, and O. Ismail, “Optimized decentralized swarm communication algorithms for efficient task allocation and
power consumption in swarm robotics,” Robotics, vol. 13, no. 5, Apr. 2024, doi: 10.3390/robotics13050066.
[29] F. J. Murray, “Perturbation theory and Lie algebras,” Journal of Mathematical Physics, vol. 3, no. 3, pp. 451–468, May 1962, doi:
10.1063/1.1724245.
[30] F. C. Park, “Computational aspects of the product-of-exponentials formula for robot kinematics,” IEEE Transactions on
Automatic Control, vol. 39, no. 3, pp. 643–647, Mar. 1994, doi: 10.1109/9.280779.
[31] H. E. Elsayed, N. S. Tawfik, O. Shalash, and O. Ismail, “Enhancing human emotion classification in human-robot interaction,” in
2024 International Conference on Machine Intelligence and Smart Innovation, ICMISI 2024 - Proceedings, May 2024, pp. 19–24.
doi: 10.1109/ICMISI61517.2024.10580152.
[32] I. Chen and G. Yang, “Kinematic calibration of modular reconfigurable robots using product‐of‐exponentials formula,” Journal of
Robotic Systems, vol. 14, no. 11, pp. 807–821, Nov. 1997, doi: 10.1002/(sici)1097-4563(199711)14:11<807::aid-rob4>3.3.co;2-3.
[33] C. He, S. Wang, Y. Xing, and X. Wang, “Kinematics analysis of the coupled tendon-driven robot based on the product-of-
exponentials formula,” Mechanism and Machine Theory, vol. 60, pp. 90–111, Feb. 2013, doi:
10.1016/j.mechmachtheory.2012.10.002.
[34] J. Yang and Z. Jason Geng, “Closed form forward kinematics solution to a class of hexapod robots,” IEEE Transactions on
Robotics and Automation, vol. 14, no. 3, pp. 503–508, Jun. 1998, doi: 10.1109/70.678460.
[35] M. A. Ali, H. A. Park, and C. S. G. Lee, “Closed-form inverse kinematic joint solution for humanoid robots,” in IEEE/RSJ 2010
International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings, Oct. 2010, pp. 704–709. doi:
10.1109/IROS.2010.5649842.
[36] Y. Liu, M. Kong, N. Wan, and P. Ben-Tzvi, “A geometric approach to obtain the closed-form forward kinematics of H4 parallel
robot,” Journal of Mechanisms and Robotics, vol. 10, no. 5, Jul. 2018, doi: 10.1115/1.4040703.
[37] S. Tejomurtula and S. Kak, “Inverse kinematics in robotics using neural networks,” Information sciences, vol. 116, no. 2, pp.
147–164, Jan. 1999, doi: 10.1016/S0020-0255(98)10098-1.
[38] J. A. Abdor-Sierra, E. A. Merchán-Cruz, and R. G. Rodríguez-Cañizo, “A comparative analysis of metaheuristic algorithms for
solving the inverse kinematics of robot manipulators,” Results in Engineering, vol. 16, Dec. 2022, doi:
10.1016/j.rineng.2022.100597.
[39] A.-V. Duka, “Neural network based inverse kinematics solution for trajectory tracking of a robotic arm,” Procedia Technology,
vol. 12, pp. 20–27, 2014, doi: 10.1016/j.protcy.2013.12.451.
[40] H. Hadfield, L. Wei, and J. Lasenby, “The forward and inverse kinematics of a delta robot,” in Lecture Notes in Computer Science
(including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12221 LNCS, Springer
International Publishing, 2020, pp. 447–458. doi: 10.1007/978-3-030-61864-3_38.
[41] D. R. Parhi, B. B. V. L. Deepak, and A. Amrit, “Forward and inverse kinematic models for an articulated robotic manipulator,”
International Journal of Artificial Intelligence and Computational Research, vol. 4, pp. 103–109, 2013.
[42] A. El-Sherbiny, M. A. Elhosseini, and A. Y. Haikal, “A comparative study of soft computing methods to solve inverse kinematics
problem,” Ain Shams Engineering Journal, vol. 9, no. 4, pp. 2535–2548, Dec. 2018, doi: 10.1016/j.asej.2017.08.001.
[43] M. Meredith and S. Maddock, “Real-time inverse kinematics: The return of the Jacobian,” Tech. Rep. CS-04-06, Department of
Computer Science, pp. 1–15, 2004.
[44] A. Aristidou and J. Lasenby, “Inverse Kinematics: a review of existing techniques and introduction of a new fast iterative solver,”
The Lancet, vol. 158, no. 4071, 2009.
[45] R. C. O. Jesus, L. Molina, E. A. N. Carvalho, and E. O. Freire, “Singularity-free inverse kinematics with joint prioritization for
manipulators,” Journal of Control, Automation and Electrical Systems, vol. 33, no. 3, pp. 1022–1031, Jan. 2022, doi:
10.1007/s40313-021-00860-4.
[46] A. Volinski, Y. Zaidel, A. Shalumov, T. DeWolf, L. Supic, and E. Ezra Tsur, “Data-driven artificial and spiking neural networks
for inverse kinematics in neurorobotics,” Patterns, vol. 3, no. 1, Jan. 2022, doi: 10.1016/j.patter.2021.100391.
[47] B. Kenwright, “Inverse kinematics – cyclic coordinate descent (CCD),” Journal of Graphics Tools, vol. 16, no. 4, pp. 177–217,
Oct. 2012, doi: 10.1080/2165347x.2013.823362.
[48] G. Tevatia and S. Schaal, “Inverse kinematics for humanoid robots,” in Proceedings-IEEE International Conference on Robotics
and Automation, 2000, vol. 1, pp. 294–299. doi: 10.1109/ROBOT.2000.844073.
[49] S. Ojha, K. Leodler, and T. H. Wu, “Singularity-free inverse kinematics by cyclic coordinate descent of a 6 DOF robotic
manipulator,” in 2023 20th International Conference on Ubiquitous Robots, UR 2023, Jun. 2023, pp. 735–740. doi:
10.1109/UR57808.2023.10202406.
[50] H. Wu, J. Yu, J. Pan, G. Li, and X. Pei, “FBCCD: a forward and backward cyclic iterative solver for the inverse kinematics of
continuum robot,” in Lecture Notes in Mechanical Engineering, Springer Nature Singapore, 2023, pp. 329–345. doi:
10.1007/978-981-19-9398-5_18.
[51] J. Te Lin, C. Girerd, J. Yan, J. T. Hwang, and T. K. Morimoto, “A generalized framework for concentric tube robot design using
gradient-based optimization,” IEEE Transactions on Robotics, vol. 38, no. 6, pp. 3774–3791, Dec. 2022, doi:
10.1109/TRO.2022.3180627.
[52] S. Wild, T. Zeng, A. Mohammad, J. Billingham, D. Axinte, and X. Dong, “Efficient and scalable inverse kinematics for
continuum robots,” IEEE Robotics and Automation Letters, vol. 9, no. 1, pp. 375–381, Jan. 2024, doi: 10.1109/LRA.2023.3331291.

IAES Int J Rob & Autom, Vol. 14, No. 1, March 2025: 1-10
IAES Int J Rob & Autom ISSN: 2722-2586  9

[53] D. Wu, W. Zhang, M. Qin, and B. Xie, “Interval search genetic algorithm based on trajectory to solve inverse kinematics of
redundant manipulators and its application,” in Proceedings - IEEE International Conference on Robotics and Automation, May
2020, vol. 10, pp. 7088–7094. doi: 10.1109/ICRA40945.2020.9196890.
[54] X. Wang, X. Liu, L. Chen, and H. Hu, “Deep-learning damped least squares method for inverse kinematics of redundant robots,”
Measurement: Journal of the International Measurement Confederation, vol. 171, Feb. 2021, doi:
10.1016/j.measurement.2020.108821.
[55] M. Ahmad, N. Kumar, and R. Kumari, “A hybrid genetic algorithm approach to solve inverse kinematics of a mechanical
manipulator,” International Journal of Scientific and Technology Research, vol. 8, no. 9, pp. 1772–1782, 2019.
[56] C. T. Lee and J. Y. Chang, “A workspace-analysis-based genetic algorithm for solving inverse kinematics of a multi-fingered
anthropomorphic hand,” Applied Sciences (Switzerland), vol. 11, no. 6, Mar. 2021, doi: 10.3390/app11062668.
[57] “Jupiter2,” Lattel Robotics - Robots. https://lattelrobotics.com/home/jupiter-robot (accessed Aug. 02, 2024).
[58] E. Lazarevska, “A neuro-fuzzy model of the inverse kinematics of a 4 DOF robotic arm,” in Proceedings - 2012 14th
International Conference on Modelling and Simulation, UKSim 2012, Mar. 2012, pp. 306–311. doi: 10.1109/UKSim.2012.51.
[59] H. Danaci, L. A. Nguyen, T. L. Harman, and M. Pagan, “Inverse kinematics for serial robot manipulators by particle swarm
optimization and POSIX threads implementation,” Applied Sciences (Switzerland), vol. 13, no. 7, p. 4515, Apr. 2023, doi:
10.3390/app13074515.

BIOGRAPHIES OF AUTHORS

Omar Shalash is an assistant professor at the Arab Academy for Science,

Technology, and Maritime Transport at the College of Artificial Intelligence in Egypt and
specializes in robotics. He holds a Ph.D. in Biomedical Engineering from the University of
Strathclyde, where he developed an autonomous robotic machine for knee arthroplasty. His
research, which includes publications on swarm robotics and electric vehicle battery
management, demonstrates his expertise in both theoretical and applied aspects of computer
engineering, robotics, and AI. Dr. Shalash is committed to advancing artificial intelligence and
robotics through both his teaching and numerous supervised research projects. He can be
contacted at [email protected].

Adham Sakr is an AI and robotics engineer with a solid background in

autonomous systems, machine learning, and intelligent robotics. Proven experience in creating
practical solutions for challenging environments, including autonomous driving and service
robotics. Currently exploring how to integrate other technologies with AI technologies to
enhance automation and efficiency in real-world scenarios. He committed to advancing the
field through innovative projects and impactful applications. He can be contacted at
[email protected].

Yasser Salem is an AI developer and robotics engineer with expertise in smart

systems, autonomous vehicles, and robotics design. Currently pursuing my master’s at
Breitenberg Technical University (BTU), experienced with hands-on projects in AI-driven
perception and automation. He is passionate about advancing robotics and AI to create
innovative solutions. He can be contacted at [email protected].

Ahmed Abdelhadi is an AI and robotics engineer with extensive experience in

autonomous systems, machine learning, and computer vision. Skilled in developing advanced
robotics solutions, including autonomous vehicles and home assistance robots. Currently
focused on real-world applications of AI and robotics, including hazardous environment
automation. He is passionate about leveraging cutting-edge technologies to drive innovation in
AI-driven automation. He can be contacted at [email protected].

Position and orientation analysis of Jupiter robot arm for navigation stability (Omar Shalash)
10  ISSN: 2722-2586

HossamEldin Elsayed is a teaching assistant at the College of Artificial

Intelligence, Arab Academy for Science, Engineering, Alamein Campus. He completed his
bachelor’s degree in computer engineering from the Arab Academy for Science, Technology,
and Maritime Transport in 2020. Recognizing his passion for the field, HossamEldin continued
his academic journey at the same institution, where he is currently pursuing a master’s degree
in computer engineering. He can be contacted at [email protected].

Ahmed El-Shaer is an associate professor at the College of Artificial

Intelligence, Arab Academy for Science, Engineering, Alamein Campus. His work mainly
with fabrication, supercapacitors, and electronic measurements. He can be contacted at
[email protected].

IAES Int J Rob & Autom, Vol. 14, No. 1, March 2025: 1-10