Indonesian J our nal of Electrical Engineering and Computer Science V ol. 40, No. 3, December 2025, pp. 1298 ∼ 1307 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v40.i3.pp1298-1307 ❒ 1298 P otential eld-based appr oaches f or nanobotics in drug deli v ery Leonardo Kamajaya 1 , Indrazno Siradjuddin 1 , Gillang Al Azhar 1 , Fitri 1 , Agwin F ahmi F ahanani 2 1 Department of Electronics Engineering, F aculty of Electrical Engineering, State Polytechnic of Malang, Malang, Indonesia 2 Department of Ph ysiology , F aculty of Medicine, Uni v ersitas Bra wijaya, Malang, Indonesia Article Inf o Article history: Recei v ed Mar 4, 2025 Re vised Oct 17, 2025 Accepted No v 23, 2025 K eyw ords: Drug deli v ery Local minima Nanorobotics Obstacle a v oidance Potential eld ABSTRA CT Nanorobotics has transformed tar geted drug deli v ery by enhancing therapeutic ef cac y , minimizing of f-tar get ef fects, and increasing precision. Ho we v er , na v- ig ating comple x biological en vironments is challenging. In the eld of macro- scopic robotics, potential eld (PF)-based approaches that utilize attracti v e and repulsi v e virtual forces pro vide a promising frame w ork that can be applied to path planning for nanorobots. Thi s study modies PF algorithms for nanorobotic na vig ation to address challenges such as a v oiding dynamic obstacles, escaping local minima, and optimizing trajectories in real time. W e e v aluated the mo v e- ment of the nanorobot through simulations under static and dynamic conditions for the tar gets and obstacles. The results demonstrate that nanorobotics with h y- brid PF methodologies enhance na vig ation performance, enabling nanorobots to successfully na vig ate through biological barriers and ef ciently reach their tar - get locati ons. This w ork is a signicant step to w ards intelligent and autonomous nanorobotic drug deli v ery systems and contrib utes to practical biomedical appli- cations. This is an open access article under the CC BY -SA license . Corresponding A uthor: Agwin F ahmi F ahanani Department of Ph ysiology , F aculty of Medicine, Uni v ersitas Bra wijaya, Malang, Indonesia Jl. V eteran No. 10-11, Malang, East Ja v a, Indonesia Email: agwinf ahmi@ub .ac.id 1. INTR ODUCTION Nanorobotics has re v olutionized tar geted drug deli v ery by of fering greater accurac y in comple x bio- logical en vironments. Con v entional methods suf fer from systemic toxicity , limited precision, and inef cient tra v ersal of ph ysiological barriers. In contrast, nanorobots of fer precise treatment del i v ery , minimize unin- tended ef fects, and enhance therapeutic outcomes [ 1 ]-[ 3 ]. Despite these technological de v elopments, ef fec- ti v ely na vig ating comple x and dynamic biological en vironments remains challenging. The application of potential eld (PF) techniques, which ha v e been widely utilized in path planning for lar ge-scale robotics, presents a promising frame w ork for guiding nanorobots [ 4 ], [ 5 ]. These algorithms simulate attracti v e and repulsi v e forces to f acilitate real-time trajectory adjustment. Ho we v er , con v entional PF techniques are susceptible to limitations, such as local minima and insuf cient adaptability to dynamic tar gets [ 6 ]. Despite recent adv ancements in h ybrid PF approaches that address specic limitations, further enhancements are necessary to ensure their ef fecti v e implementation at the nanoscale. Unlik e traditi onal PF methods tailored for macroscopic robots [ 4 ], [ 6 ], [ 7 ], this study presents a nanoscale-adapted PF frame w ork that incorporates adapti v e kinematics, real-time trajectory adjustments, and h ybrid control mechanisms for na vig ating biological microen vironments. Building on recent adv ancements J ournal homepage: http://ijeecs.iaescor e .com Evaluation Warning : The document was created with Spire.PDF for Python.
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752 ❒ 1299 in articial potential eld (APF) models, such as the bacterial PF approach [ 5 ], the proposed method e xtends these principles to nanoscale drug deli v ery . It does so by inte grating dynamic force eld modeling capable of managing both static and dynamic obstacles with minimal path de viation. This nanoscale adaptation addresses challenges related to viscous drag, Bro wnian motion, and en vironmental uncertainties, thereby impro ving tra- jectory con v er gence, stability , and responsi v eness. Ov erall, the approach signicantly adv ances the practical implementation of autonomous nanorobotic na vig ation for precise and ef cient tar geted drug deli v ery . 2. THEORETICAL APPR O A CH 2.1. P otential eld PF algorithms create virtual force elds to direct robots to w ards tar gets (attracti v e forces, F a ) and a w ay from obstacles (repulsi v e forces, F r ) [ 8 ]-[ 10 ]. The cumulati v e force ( F x ) applied on a robot is articulated as: F x = F a + F r (2..1) In a fundamental PF scenario, the robot percei v es barriers as repulsi v e PFs and the objecti v e as an appealing eld. Let x = ( x, y ) T signify the robot’ s position, thereafter the area in which the robot op e rates is referred to as: U( x ) , where: U( x ) = U a ( x ) + U r ( x ) (2..2) and U a ( x ) denotes the attracti v e eld and U r ( x ) e xpresses the repulsi v e eld. It may also be e xpressed as: U( x ) = U a ( x ) + N X n =1 U ( n ) r ( x ) (2..3) where N represents the total quantity of barriers within the nanorobot’ s operational en vironment. The force propelling the nanorobot is the ne g ati v e gradient of the PF , that is: F x = −∇ x U a ( x ) | {z } F a (x) −∇ x U r ( x ) | {z } F r (x) (2..4) where F a (x) = ( F a,x , F a,y ) T is the attracti v e force v ector and F r (x) = ( F r ,x , F r ,y ) T is the repulsi v e force v ector . Con v entional PF methods encounter challenges such as local minima, tar get inaccessibility , and li m- ited dynamic obstacle-a v oidance. Researchers ha v e addressed these issues by implementing speed-dependent elds, simulated annealing, and h ybrid PF global planners [ 8 ], [ 11 ], [ 12 ], thereby enhancing na vig ation in high-density en vironments. 2.2. Nanor obotics-dri v en drug deli v ery systems Nanorobotics-dri v en drug deli v ery systems represent an i n no v ati v e approach for precise medic ation. These nanorobot de vices, typically ranging from 1 to 100 nm in size, are capable of tra v ersing the human body to deli v er pharmaceuticals directly to tar geted sites [ 13 ], [ 14 ]. The y emplo y specialized tar geting mechanisms, such as lig and-receptor interactions, magnetic guidance, or enzymatic processes, to concentrate on specic problem areas, thereby minim izing side ef fects and enhancing therapeutic ef cac y [ 15 ], [ 16 ]. Compared to con v entional methods, this approach of fers superior control o v er pharm acokinetics, necessitates lo wer dosages, and f acilitates patient adherence to treatment re gimens [ 17 ], [ 18 ]. A nanorobot consists of three primary components, such as propulsion system, drug storage compa rt- ment, and na vig ation unit. Propulsion may be achie v ed through magnetic, chemical, or enzymatic means or a combination thereof, enabling the robot to na vig ate through viscous bodily uids. Na vig ation is directed by sophisticated algorithms that assist a robot in circumv enting obstacles and maintaining its trajectory . These features are essential for the precise deli v ery of drugs within the body . Although there are some challenges in this area lik e ensuring biocompatibility , pre v enting i mmune reactions, maintaining stability inside the body , and achie ving mass production, adv ancements are still being made [ 2 ], [ 19 ]. Inno v ati v e s olutions lik e biode gradable coatings, lipid carriers, and enzyme surf aces ha v e impro v ed the inte gration of these robots with biological systems. Adv ances in 3D printing, self-assembly , and sw arm coordination ha v e made these systems increasingly viable for medical applications [ 20 ], [ 21 ]. P otential eld-based appr oac hes for nanobotics in drug delivery (F ahanani) Evaluation Warning : The document was created with Spire.PDF for Python.
1300 ❒ ISSN: 2502-4752 3. PR OPOSED METHOD 3.1. Kinematic of nanor obotics Nanorobots rely on re gulated kinematic motion to achie v e precise drug deli v ery and tar geting. Their mo v ement is inuenced by ambient conditions, propulsion mechanisms and real-time feedback systems. Kine- matic modeling of nanorobots in v olv es characterizing their motion parameters, ener gy ef cienc y , and adapta- tion within dynamic biological en vironments [ 22 ], [ 23 ]. The mobility of nanorobots is characterized by the interactions between v elocity and force. The propulsion force F mag netic for magnetic-dri v en nanorobots is obtained from e xternal eld gradients. F mag netic = ∇ ( m · B ) (3..1) Where m is the magnetic moment and B is the applied eld. Nanorobotic motion is inuenced by v arious f actors, including ambient conditions and uid viscosity , which is a k e y f actor in determining drag forces ( F dr ag ). F dr ag = 6 π η r v (3..2) where η is uid viscosity , r is nanorobot radius, and v is v elocity . 3.2. P otential eld algorithm 3.2.1. Attracti v e potential eld The attracti v e PF is dened as the quadratic of the distance between the nanorobot position and the objecti v e position d 2 a . The attracti v e PF becomes more pronounced as the nanorobot’ s position de viates from the objecti v e position [ 24 ], [ 25 ]. The enticing PF function can be e xpressed as: U a ( x ) = 1 2 k a d 2 a (3..3) = 1 2 k a ∥ ( x ∗ − x ) ∥ 2 (3..4) where k a is the attracti v e PF constant and x ∗ is the goal postion. Assuming that the goal position is static then: −∇ x U a ( x ) = − ∂ U a ( x ) ∂ x = k a ∥ ( x ∗ − x ) ∥ (3..5) therefore the attracti v e force is: F a ( x ) = k a ∥ ( x ∗ − x ) ∥ (3..6) where d a = ∥ ( x ∗ − x ) ∥ = p ( x ∗ − x ) 2 + ( y ∗ − y ) 2 is the Euclidean distance between the goal postion x ∗ = ( x ∗ , y ∗ ) T and the nanorobot position x = ( x, y ) T . 3.2.2. Repulsi v e potential eld In contrast to the attracti v e PF , the repulsi v e PF is more potent when the nanorobot is situated in close proximity to the obstacle. An y impediment that is located at a greater distance will not lik ely contrib ute to the total PF . Consequently , if the nanorobot’ s distance from the n -th obstacle, d ( n ) o , e xceeds a threshold v alue τ , the respected PF is not considered, resulting in U ( n ) r (x) = 0 . Alternately , the repulsi v e PF is proportional to the in v erse of the distance between the obstacle and the automaton [ 24 ], [ 25 ]. Let determine the repulsi v e function as: U ( n ) r (x) = 1 2 k r 1 d ( n ) o 2 if d ( n ) o ≤ τ 0 if d ( n ) o > τ (3..7) where d ( n ) o = ∥ ( x n o − x ) ∥ = p ( x n o − x ) 2 + ( y n o − y ) 2 is the Euclidean distance between the nanorobot posi- tion x = ( x, y ) T and the n -th obstacle position x (n) o = ( x ( n ) o , y ( n ) o ) T . Assuming that each obstacle possesses the same repulsi v e PF characteristic, the repulsi v e PF constant is represented by k r . The repulsi v e force or ne g ati v e gradient, is calculated as: −∇ x U ( n ) r (x) = − ∂ U ( n ) r (x) ∂ x = F ( n ) r (x) = k r d ( n ) o 3 if d ( n ) o ≤ τ 0 if d ( n ) o > τ (3..8) Indonesian J Elec Eng & Comp Sci, V ol. 40, No. 3, December 2025: 1298–1307 Evaluation Warning : The document was created with Spire.PDF for Python.
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752 ❒ 1301 3.3. Nanor obot contr ol f or obstacle a v oidance Nanorobots can autonomously na vig ate intricate microstr uctures by combini ng vision-based control systems, magnetic eld manipulation, adapti v e motion control, reinforcement learning, and adv anced path- planning algorithms. Nanorobots can identify and e v ade obstructions in real time o wing t o the collecti v e application of these technologies, which guarantees precise and ef cient mobility in dynamically changing en vironments. Prior to further de v eloping the control rule for robot motion to e v ade obstacles, it is essential to decompose all forces, both attracting and repulsi v e, into their projections along the x-axis and y-axis [ 24 ], [ 25 ]. Consequently , the breakdo wn of the attracti v e force is calculated as: F a,x = −∇ x U a ( x ) (3..9) = k a ( x ∗ − x ) (3..10) note that d a = p ( x ∗ − x ) 2 + ( y ∗ − y ) 2 and F a,x is the attracti v e force int the x -axis direction. Moreo v er , the attracti v e force in y -axis direction is denoted as: F a,y = k a ( y ∗ − y ) (3..11) the projections of the repulsi v e force along the x -axis and y -axis are as follo ws (note: the deri v ation considers just the case when d ( n ) o ≤ τ ): F ( n ) r ,x = −∇ x U ( n ) r ( x ) (3..12) = k r ( x ( n ) o − x ) d ( n ) o 4 (3..13) where F ( n ) r ,x represents the repulsi v e force e x erted by the n -th obstacle along the x -axis. The identical deduction is applicable to ascertain the repulsi v e force in the y -axis direction, where: F ( n ) r ,y = k r ( y ( n ) o − y ) d ( n ) o 4 (3..14) therefore, the total force F(x) = ( F x , F y ) T can be calculated as: F x = F a,x (x) + N X n =1 F ( n ) r ,x (x) (3..15) F y = F a,y (x) + N X n =1 F ( n ) r ,y (x) (3..16) nally , the force direction can be computed as: θ = arctan F y F x (3..17) therefore, the v elocity control signal v = ( v x , v y ) T is determined by the total PF force in the simplest nanorobot control algorithm. v x = F x (3..18) v y = F y (3..19) Then the nanorobot position can be updated by: x ( k ) = x ( k − 1) + v x ∆ t (3..20) y ( k ) = y ( k − 1) + v y ∆ t (3..21) where ∆ t is the sampling time, x ( k ) and y ( k ) are the nanorobot x and y coordinates at k -th time instant. P otential eld-based appr oac hes for nanobotics in drug delivery (F ahanani) Evaluation Warning : The document was created with Spire.PDF for Python.
1302 ❒ ISSN: 2502-4752 The k e y design parameters in this w ork include the attracti v e constant k a , the repulsi v e constant k r , and the repulsi v e threshold distance τ . These parameters were chosen through empirical tuning and theoretical stability considerations to ensure smooth con v er gence without oscillation. In practice, a high k a increases con v er gence speed b ut may lead to o v ershooting near the tar get, while a high k r impro v es obstacle a v oidance b ut can result in unnecessary detours. The threshold τ w as determined based on the nanorobot size and the e xpected biological barrier distance, typically 2–5 times the nanorobot diameter . Optimal parameters were obtained by iterati v e tuning across four simulation states until the nanorobot reached the tar get with minimal path error . 4. RESUL TS AND DISCUSSION Algorithm testing w as conducted using a series of simulations that modelled four distinct conditions. I n the initial simulation, the nanorobot na vig ated an en vironment with stati c obstacles and tar gets. The second simulation incorporated static obstacles with dynamic tar gets. In the third, the obstacles were dynamic while the tar get remained static. The fourth simulation inte grated both dynamic obstacles and dynamic tar gets. In all instances, the F a and F r v alues were maintained constant. The coordinates of the obstacles and tar gets were modied to represent static or dynamic conditions. The simulation results are presented belo w . Figure 1 illustrate general o wchart to e xplain the simulation process performed on the nanorobot motion study . Figure 1. General o wchart to e xplain the simulation process performed on the nanorobot motion study 4.1. Nanor obot motion with a static tar get and static obstacle The simulation w as conducted by performing tw o trials, with a x ed tar get coordinate point x = 10, y = 5, and a x ed starting point with coordinate x = 0, y = 0. Additionally , four obstacles were positioned in tw o dif ferent location schemes. The positions of the obstacles can be seen in T able 1 . Figure 2 sho ws test results of nanorobot mo v ement with rst condition. Based on the simulation results sho wn in Figure 2(a) and Figure 2(b), it can be observ ed that the nanorobot can maneuv er to a v oid obstacles to reach the tar get coordinate point. It is e vident that when the nanorobot approaches an obsta- cle and enters the repulsi v e re gion of the obstacle, the nanorobot can mo v e to w ards a lo wer repulsi v e area (the magnitude of the repulsi v e force is indicated by the darkness of the blue color around the obstacle area). Indonesian J Elec Eng & Comp Sci, V ol. 40, No. 3, December 2025: 1298–1307 Evaluation Warning : The document was created with Spire.PDF for Python.
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752 ❒ 1303 4.2. Nanor obot motion with a dynamic tar get and static obstacle The simulation w as conducted by performing tw o tests, with a x ed starting point with coordinate x = 0, y = 0. Additionally , tw o dif ferent dynamic tar get coordinate points, which i s: i) mo ving tar get which mo ving from x = 8.0, y = 3.0 into x = 6.0, y = 8.0 and ii) mo ving tar get which mo ving from x = 10.0, y = 5.0 into x = 10.0, y = 0.0. Moreo v er , four obstacles were positioned i n tw o dif ferent locations for each scheme. The positions of the obstacles can be seen in T able 1 . T able 1. Obstacle conguration for each scheme under condition I and II Scheme 1 Scheme 2 Obstacle x y x y 1 2.0 2.0 2.0 2.0 2 3.0 1.0 3.0 1.0 3 7.0 2.5 5.0 4.0 4 7.5 1.5 6.0 4.5 Figure 2 sho ws test results of nanorobot mo v ement with second condition. Based on the sim ulation results sho wn in Figure 2(c) and Figure 2(d), it can be observ ed that the nanorobot can maneuv er to a v oid obstacles to reach the tar get coordinate, which is also in motion into the nal position. In the pre vious results, the nanorobot also a v oids obstacles based on the repulsi v e force and the attracti v e force from the tar get, and the attracti v e force from the tar get v aries in v alue because the tar get destinations of the nanorobot are also in motion. It enables the nanorobot to maneuv er with dif ferent v ariations, adapting to the mo v ements of the tar get being pursued. The maneuv er v ariations generated also depend on the closest trajectory of the nanorobot to the mo ving tar get. (a) (b) (c) (d) Figure 2. Nanorobot mo v ement test results under dif ferent simulation conditions (a) and (b) Condition I (Static tar get and static obstacles) for Scheme 1 and Scheme 2. (c) and (d) Condition II (Dynamic tar get and static obstacles) for Scheme 1 and Scheme 2 4.3. Nanor obot motion with a static tar get and dynamic obstacle The simulation w as conducted by performing tw o trials, with a x ed tar get coordinate point x = 8, y = 6, x ed starting point with coordinate x = 0, y = 0, and tw o obstacles were positioned in tw o dif ferent locations that mo v ed into a specic location for each scheme. The positions of the obstacles can be seen in T able 2 . T able 2. Obstacle conguration for each scheme under condition III Scheme 1 Scheme 2 Initial pose Final pose Initial pose Final pose Obstacle x y x y x y x y 1 2.0 2.0 3.0 -1.0 2.0 1.0 6.5 4.0 2 8.0 5.0 5.5 4.5 7.0 5.0 6.0 2.0 Figure 3 sho ws test results of nanorobot mo v ement with second condition. Based on the sim ula- tion results sho wn in Figure 3(a) and Figure 3(b), it can be observ ed that the nanorobot can maneuv er to a v oid obstacles to reach the tar get coordinate e v en if the obstacles are also in motion. The nanorobot can re- spond to mo ving obstacles, as demonstrated when the nanorobot encounters a suddenly appearing obstacle. P otential eld-based appr oac hes for nanobotics in drug delivery (F ahanani) Evaluation Warning : The document was created with Spire.PDF for Python.
1304 ❒ ISSN: 2502-4752 The nanorobot can maneuv er to alter its trajectory to a v oid collision with the obstacle. In the gure, the trajec- tory of the obstacle is sho wn in red, while the nanorobot’ s trajectory is sho wn in black. At specic points in the gure, mark ed as collisi on points , it represents the locations where the nanorobot encounters the obstacle. Ho we v er , before a collision occurs, the nanorobot maneuv ers to a v oid the obstacle while mo ving to w ards the predetermined tar get coordinate point. As seen in both results gures, the nanorobot can instantly react to the obstacles motion by maneuv ering. 4.4. Nanor obot motion with a dynamic tar get and dynamic obstacle The simulation w as conducted by performing tw o trials, with a x ed st arting point with coordinate x = 0, y = 0. Additionally , tw o dif ferent dynamic tar get coordinate points, which is: i) mo ving tar get which mo ving from x = 8.0, y = 0.0 into x = 8.0, y = 7.0 and ii) mo ving tar get which mo ving from x = 8.5, y = 4.0 into x = 2.0, y = 8.0. Furthermore, tw o up to three obstacles were positioned in dif ferent l ocations that mo v ed into a specic location for each scheme. The positions of the obstacles can be seen in T able 3 . Figure 3 sho ws test results of nanorobot mo v ement with second condition. Based on the simulation results sho wn in Figure 3(c) and Figure 3(d), it can be observ ed that the nanorobot can maneuv er to a v oid obstacles to reach the tar get coordinate, e v en if the obstacles are also in motion. The results indica te that the nanorobot can mo v e to w ards a mo ving tar get by adjusting its trajectory to chase the mo v ement of that tar get point. During the pursuit of the tar get point’ s mo v ement, the nanorobot encounters mo ving obstacles that can potentially lead to collisions . Ho we v er , from the e x ecuted simulation, the nanorobot can maneuv er to a v oid suddenly appearing obstacles while maintaining its mo v ement to w ards the tar get destination. Therefore, through these maneuv ers, it is e vident that the nanorobot can mo v e agilely to a v oid mo ving obstacles while still preserving its direction to w ards the mo ving tar get point. T able 3. Obstacle conguration for each scheme under condition IV Scheme 1 Scheme 2 Initial pose Final pose Initial pose Final pose Obstacle x y x y x y x y 1 2.0 1.0 7.0 0.75 2.0 2.0 4.5 2.0 2 7.5 5.0 6.0 2.0 8.0 0.0 5.0 2.5 3 - - - - 4.0 4.0 5.5 5.5 (a) (b) (c) (d) Figure 3. Nanorobot mo v ement test results under dif ferent simulation conditions (a) and (b) Condition III (Static tar get and Dynamic obstacles) for Scheme 1 and Scheme 2. (c) and (d) Condition IV (Dynamic tar get and Dynamic obstacles) for Scheme 1 and Scheme 2 5. CONCLUSION This study has illustrated the ef cac y of PF-based methodologies in impro ving the autonomous na v- ig ation of nanorobots for medication deli v ery purposes. Utilizing appealing and repulsi v e virtual forces, nanorobots may adeptly na vig ate intricate biol og i cal settings while reducing collision hazards. The simula- tion outcomes v alidate that PF techniques f acilitate accurate path planni n g, e v en in dynamic en vironments with mobile barriers and tar gets. Con v entional PF methods encounter obstacles, including local minima and re- stricted adaptation to rapidly changing ph ysiological conditions. Hybrid PF approaches that inte grate global optimization, reinforcement learning, and v elocity-dependent modications present viable options to address these limitations. Future research must prioritize e xperimental v alidation in authentic biological settings and Indonesian J Elec Eng & Comp Sci, V ol. 40, No. 3, December 2025: 1298–1307 Evaluation Warning : The document was created with Spire.PDF for Python.
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752 ❒ 1305 the inte gration of multi-modal control mechanisms to impro v e na vig ation ef cienc y . This study’ s ndings en- hance the de v elopment of intelligent nanorobotic system s for tar geted drug deli v ery , f acilitating more ef cient and least intrusi v e therapeutic interv entions. FUNDING INFORMA TION Authors state no funding in v olv ed. A UTHOR CONTRIB UTIONS ST A TEMENT This journal uses the Contrib utor Roles T axonomy (CRediT) to recognize indi vidual author contrib u- tions, reduce authorship disputes, and f acilitate collaboration. Name of Author C M So V a F o I R D O E V i Su P Fu Leonardo Kamajaya √ √ √ √ √ √ √ √ Indrazno Siradjuddin √ √ √ √ √ √ Gillang Al Azhar √ √ √ √ √ √ √ √ √ √ Fitri √ √ √ √ √ Agwin F ahmi F ahanani √ √ √ √ C : C onceptualization I : I n v estig ation V i : V i sualization M : M ethodology R : R esources Su : Su pervision So : So ftw are D : D ata Curation P : P roject administration V a : V a lidation O : Writing - O riginal Draft Fu : Fu nding acquisition F o : F o rmal analysis E : Writing - Re vie w & E diting CONFLICT OF INTEREST ST A TEMENT Authors state no conict of interest. D A T A A V AILABILITY The authors conrm that all data supporting the ndings of this study are a v ailable within the article. REFERENCES [1] K. Xu and J. Zhou, “Drug deliery for micro-/nanorobots: progress and challenges, ” Curr ent Nanoscience , v ol. 18, no. 6, pp. 690–699, 2022, doi: 10.2174/1573413718666220127123038. [2] T . K. Ramchandani, “ A brief re vie w on nanorobotics, ” International J ournal for Resear c h in Applied Science and Engineering T ec hnolo gy , v ol. 11, no. 12, pp. 1301–1306, 2023, doi: 10.22214/ijraset.2023.57591. [3] H. Zhou, C. C. Mayor g a-Martinez, S. P an ´ e, L. Zhang, and M. Pumera, “Magnetically dri v en micro and nanorobots, ” Chemical Re vie ws , v ol. 121, no. 8, pp. 4999–5041, 2021, doi: 10.1021/acs.chemre v .0c01234. [4] Y . Hanli, L. Qiliang, and Z. W eizhen, “Robot obstacle a v oidance based on impro v ed articial potential eld method, ” in 10th International Confer ence on Contr ol, A utomation and Information Sciences, ICCAIS 2021 - Pr oceedings , 2021, pp. 769–775, doi: 10.1109/ICCAIS52680.2021.9624644. [5] O. Montiel, U. Orozco-Rosas, and R. Sep ´ ulv eda, “P ath planning for mobile robots using bacterial potential eld for a v oiding static and dynamic obstacles, ” Expert Systems with Applicati ons , v ol. 42, no. 12, pp. 5177–5191, 2015, doi: 10.1016/j.esw a.2015.02.033. [6] S. Amoh, D. Ito, and T . Ueta, “ A method to suppress local minima for symmetrical DOPO netw orks, ” Nonlinear Theory and Its Applications, IEICE , v ol. 11, no. 4, pp. 580–589, 2020, doi: 10.1587/nolta.11.580. [7] N. He, Y . Su, J. Guo, X. F an, Z. Liu, and B. W ang, “Dynamic path planning of mobile robot based on articial potential eld, ” in Pr oceedings - 2020 International Confer ence on Intellig ent Computing and Human-Computer Inter action, ICHCI 2020 , 2020, pp. 259–264, doi: 10.1109/ICHCI51889.2020.00063. [8] H. Li, “Robotic path planning strate gy based on impro v ed articial potential eld, ” in Pr oceedings - 2020 International Confer ence on Articial Intellig ence and Computer Engineering , ICAICE 2020 , 2020, pp. 67–71, doi: 10.1109/ICAICE51518.2020.00019. [9] Y . Li, B. T ian, Y . Y ang, and C. Li, “P ath planning of robot based on articial potential eld method, ” in IEEE 6th Information T ec h- nolo gy and Mec hatr oni cs Engineering Confer ence , IT OEC 2022 , 2022, v ol. 6, pp. 91–94, doi: 10.1109/IT OEC53115.2022.9734712. [10] R . Szczepanski, A. Bereit, and T . T arcze wski, “Ef cient local path planning algorithm using articial potential eld supported by augmented reality , ” Ener gies , v ol. 14, no. 20, p. 6642, 2021, doi: 10.3390/en14206642. [11] D. Zhang, G. Zhu, and Q. Zhang, “Multi-robot motion planning: a learning-based articial potential eld solution, ” in Pr o- ceedings of the 2nd Confer ence on Fully Actuated System Theory and Appli cations, CF AST A 2023 , 2023, pp. 233–238, doi: 10.1109/CF AST A57821.2023.10243195. P otential eld-based appr oac hes for nanobotics in drug delivery (F ahanani) Evaluation Warning : The document was created with Spire.PDF for Python.
1306 ❒ ISSN: 2502-4752 [12] R. W ang, J. Guo, S. Guo, Q. Fu, and J. Xu, “Cooperati v e hunting of spherical multi-robots based on impro v ed articial poten- tial eld m ethod, ” in 2022 IEEE International Confer ence on Mec hatr onics and A utomation, ICMA 2022 , 2022, pp. 575–580, doi: 10.1109/ICMA54519.2022.9856156. [13] S. Fusco et al. , “Shape-switching microrobots for medical applications: the inuence of shape in drug deli v ery and locomotion, ” A CS Applied Materials and Interfaces , v ol. 7, no. 12, pp. 6803–6811, 2015, doi: 10.1021/acsami.5b00181. [14] A. C. Hortel ˜ ao, T . P ati ˜ no, A. Perez-Jim ´ enez, ` A. Blanco, and S. S ´ anchez, “Enzyme-po wered nanobots enhance anticancer drug deli v ery , ” Advanced Functional Materials , v ol. 28, no. 25, p. 1705086, 2018, doi: 10.1002/adfm.201705086. [15] L. Huang, F . Chen, Y . Lai, Z. Xu, and H. Y u, “Engineering nanorobots for tumor -tar get ing drug deli v ery: from dynamic control to stimuli-responsi v e strate gy , ” ChemBioChem , v ol. 22, no. 24, pp. 3369–3380, 2021, doi: 10.1002/cbic.202100347. [16] W . Y . Jeong, M. Kw on, H. E. Choi, and K. S. Kim, “Recent adv ances in transdermal drug deli v ery systems: a re vie w , ” Biomaterials Resear c h , v ol. 25, no. 1, p. 24, 2021, doi: 10.1186/s40824-021-00226-6. [17] A. M. V ar g ason, A. C. Anselmo, and S. Mitragotri, “The e v olution of commercial drug deli v ery technologies, ” Natur e Biomedical Engineering , v ol. 5, no. 9, pp. 951–967, 2021, doi: 10.1038/s41551-021-00698-w . [18] B. Zhang, L. Zhu, H. P an, and L. Cai, “Biocompa tible smart micro/nanorobots for acti v e g astrointestinal tract drug deli v ery , ” Expert Opinion on Drug Delivery , v ol. 20, no. 10, pp. 1427–1441, 2023, doi: 10.1080/17425247.2023.2270915. [19] W . Siti, H. L. T oo, T . Anderson, X. R. Liu, I. Y . Loh, and Z. W ang, “ Autonomous DN A molecular motor tailor -designed to na vig ate DN A orig ami surf ace for f ast comple x motion and adv anced nanorobotics, ” Science Advances , v ol. 9, no. 38, p. eadi8444, 2023, doi: 10.1126/SCIAD V .ADI8444. [20] F . Soto, J. W ang, R. Ahmed, and U. Demirci, “Medical micro/nanorobots in precision medi cine, ” Advanced Science , v ol. 7, no. 21, p. 2002203, 2020, doi: 10.1002/advs.202002203. [21] Z. W u, Y . Chen, D. Mukasa, O. S. P ak, and W . Gao, “Medical micro/nanorobots in comple x media, ” Chemical Society Re vie ws , v ol. 49, no. 22, pp. 8088–8112, 2020, doi: 10.1039/d0cs00309c. [22] L. Mertz, “Enhancing therapeutic deli v ery using micro- and nanorobots, ” IEEE Pulse , v ol. 14, no. 2, pp. 18–22, 2023, doi: 10.1109/MPULS.2023.3269754. [23] Z. Sun and Y . Hou, “Micro/nanorobots as acti v e deli v ery systems for biomedicine: from self-propulsion to controllable na vig ation, ” Advanced Ther apeutics , v ol. 5, no. 7, p. 2100228, 2022, doi: 10.1002/adtp.202100228. [24] I. Siradjuddin et al. , “Mobile robot with independent steering and dri ving and obstacle a v oidance using an articial potential eld, ” in AIP Confer ence Pr oceedings , 2024, v ol. 2927, no. 1, p. 40012, doi: 10.1063/5.0205228. [25] K. Nurhalim, I. S iradjuddin, and R. I. Putri, “Implementasi algoritma potential eld obstacle a v oidance pada robot beroda 4 omni wheels simetris, ” J ournal of Applied Smart Electrical Network and Systems , v ol. 3, no. 01, pp. 16–24, Jun. 2022, doi: 10.52158/jasens.v3i01.417. BIOGRAPHIES OF A UTHORS Leonardo Kamajaya is a lecturer in electronic engineering study program at State Poly- technic of Malang. He recei v ed a bachel or’ s de gree in electronic engineering at Politeknik Elektron- ika Ne geri Surabaya and a master’ s de gree in electrical engineering at Chang Gung Uni v ersity . His research interests include a biosensor , electronics systems, electronics projects based on the IoT and rene w able ener gy . He can be contacted at email: leonardo42@polinema.ac.id. Indrazno Siradjuddin recei v ed the B.E. and mast er de grees in electrical engineering from Bra wijaya Uni v ersity , Indonesia, in 2000 and from Institut T eknologi Sepuluh Nopember , Indonesia, in 2004, and the Ph.D. de gree in computing and engineering from the Uni v ersity of Ulster , U.K, in 2014. He is currently an associate professor in the Department of Electrical Engineering, State Polytechnic of Malang, Indonesia. He has more than 160 papers to his credit publis hed in refer - eed journals and conference proceedings. His research interests include intelligent control, robotics, neural netw orks, and cogniti v e modelling. He can be contacted by email: indrazno@polinema.ac.id. Gillang Al Azhar is a lecturer in the Department of Elec trical and Electronic Engineering at State Polytechnic of Malang, Indonesia. He holds a master of engi neering de gree and specializes in mobile robotics. His e xpertise includes robot operating system (R OS), C programming, Python, Arduino, and A VR microcontrollers. His research interests focus on impro ving the accurac y and performance of mobile robotic systems. He has w ork ed on implementing g-h lters t o enhance compass sensor accurac y for dif ferential dri v e robot trajectory tracking and has contrib uted to the de v elopment of data distrib ution control s ystems using R OS for mobile robots. He can be contacted at email: gillang al azhar@polinema.ac.id. Indonesian J Elec Eng & Comp Sci, V ol. 40, No. 3, December 2025: 1298–1307 Evaluation Warning : The document was created with Spire.PDF for Python.
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752 ❒ 1307 Fitri is a lecturer in electr onic engineering study program at State Polytechnic of Malang. She recei v ed bachelor’ s and master’ s de gre es at Bra wijaya Uni v ersity . Her research in- terests include a sensor , rene w able ener gy and embedded system. She can be contacted by email: tri@polinema.ac.id. Agwin F ahmi F ahanani is a biomedical engineering researcher and lecturer in medical bachelor in F aculty of Medicine, Uni v ersitas Bra wijaya, Ma lang, Indonesia. He recei v ed a bach- elor’ s de gree in electrical engineering at Uni v ersitas Bra wijaya and a master’ s de gree in electrical engineering (biomedical engineering consentration) at Institut T eknologi B andung. His research in- terests include medical image processing, biosensor , and medical de vices. He can be contacted at email: agwinf ahmi@ub .ac.id. P otential eld-based appr oac hes for nanobotics in drug delivery (F ahanani) Evaluation Warning : The document was created with Spire.PDF for Python.