7.9
CiteScore
 
3.6
Impact Factor
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
ABUNDANCE ESTIMATION IN AN ARID ENVIRONMENT
Case Study
Correspondence
Corrigendum
Editorial
Full Length Article
Invited review
Letter to the Editor
Original Article
Research Article
Retraction notice
REVIEW
Review Article
SHORT COMMUNICATION
Short review
7.2
CiteScore
3.7
Impact Factor
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
ABUNDANCE ESTIMATION IN AN ARID ENVIRONMENT
Case Study
Correspondence
Corrigendum
Editorial
Full Length Article
Invited review
Letter to the Editor
Original Article
Research Article
Retraction notice
REVIEW
Review Article
SHORT COMMUNICATION
Short review
View/Download PDF

Translate this page into:

Research Article
2026
:38;
13312025
doi:
10.25259/JKSUS_1331_2025

Navigational ambiguity in robot path-finding: Complement metric dimension of symmetric convex planar spaces

Department of Mathematics, The Islamia University of Bahawalpur, Bahawalpur, 63100, Pakistan
Department of Electrical Engineering, Prince Mohammad Bin Fahd University (PMU), P.O. Box 1664, Al Khobar 31952, Saudi Arabia

*Corresponding author: E-mail address: mimran1@pmu.edu.sa (M Imran)

Licence
This is an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-Share Alike 4.0 License, which allows others to remix, transform, and build upon the work non-commercially, as long as the author is credited and the new creations are licensed under the identical terms.

Abstract

In graph-based models of robot navigation, landmark selection plays a critical role in enabling accurate movement across a spatial environment. A metric generator ensures unique positional identification, aiding effective path-finding. Conversely, a metric resolving complement consists of those points that fail to resolve all other positions, thus representing sources of navigational ambiguity or distraction. The cardinality of the largest such set defines the complement metric dimension of the space. In this paper, we study this concept in the context of six structured symmetric convex planar (SCP) spaces, each representing a navigable environment for a robot. We explicitly construct maximum metric resolving complements and determine exact values of the complement metric dimension for each SCP space. Our results offer theoretical insights into the identification of worst-case landmark configurations in robotic path-finding systems.

Keywords

Complement metric dimension
Graph-based pathfinding
Landmark-based localization
Metric resolving complement
Robot navigation
Symmetric convex planar space
2020 MSC: 05C12, 05C75, 68T40, 68R10

1. Introduction

In autonomous robotic systems, navigation through an environment relies heavily on the ability to interpret distances from selected reference points, known as landmarks. These landmarks are modeled mathematically using the framework of graph theory, where the environment is represented as a graph space X, and distances are defined through a graph metric. A robot determines its location by comparing distances to the selected landmarks. However, when the chosen set of reference points fails to uniquely distinguish different positions, the robot may encounter navigational ambiguity. Understanding such worst-case scenarios, where the selected landmarks provide insufficient or misleading information, is of critical importance in designing robust localization systems (Khuller et al., 1996; Susilowati et al., 2019).

The graph-theoretic concept of a metric generator serves to model effective landmark placement: it is a subset of vertices such that all other vertices have unique distance representations with respect to it. Conversely, a metric resolving complement consists of a largest set of vertices that, if chosen as landmarks, fail to resolve all positions, thus acting as a source of maximum ambiguity. The cardinality of such a set is called the complement metric dimension, a concept that has applications in both robotic path-finding and intrusion detection in networks.

In this paper, we investigate the complement metric dimension of six families of symmetric convex planar (SCP) spaces, each modeling a structured geometric environment. For each space, we determine the exact value of its complement metric dimension by constructing maximal metric resolving complements and establishing corresponding upper bounds. These results provide insight into the structural limitations of landmark-based navigation in convex environments.

1.1 Metric framework and complementary resolving concepts

The concept of a metric is foundational in the theory of graph spaces and plays a pivotal role in modeling identification, navigation, and localization tasks. It has inspired extensive research into tools that uniquely identify points in a graph space. These tools not only serve to select effective landmark placements for robotic navigation but also play a critical role in other domains such as intrusion detection in networks (e.g., fire, theft) (Chartrand et al., 2000), and molecular identification for drug design and compound classification in chemistry (Chartrand et al., 2000; Salman et al., 2018).

Let X be a graph space with vertex set V(X). The metric on X is a function m:V(X)×V(X){0} defined such that m(x,y)=l, where l is the length (number of edges) of a shortest path between vertices x and y in X (Chartrand and Zhang, 2006). The diameter of X, denoted by diam(X), is the maximum metric distance between any two vertices:

diam X=max m x,y  ; x,yVX .

A fundamental notion for navigation in graph spaces is the metric generator, which supports landmark selection. Let S={s1,s2,,sk} be a subset of vertices in X. The metric code of a vertex xV(X) with respect to S is the k-tuple:

cS x= m x,s1 ,m x,s2 ,,m x,sk .

The set S is called a metric generator of X if cS(x)cS(y) for every pair of distinct vertices x,yV(X)\S. A metric generator of smallest possible size is known as a metric basis, and its cardinality is called the metric dimension of X, denoted by dim(X) (Chartrand et al., 2000; Harary and Melter, 1976; Khuller et al., 1996; Slater, 1975).

In 2019, Susilowati et al. introduced the complementary notion of a metric resolving complement (Susilowati et al. 2019). This concept identifies those vertices in a graph space that, if selected as landmarks, fail to resolve all other positions, thereby potentially misleading robotic navigation. A subset SV(X) is called a metric resolving complement (or complement metric generator) of X if there exist distinct vertices x,yV(X)\S such that cS(x)=cS(y). A resolving complement of maximum cardinality is called a complement metric basis, and its size is the complement metric dimension of X, denoted by dim ¯(X).

Research on complement metric dimension has expanded across various graph classes:

  • Complement resolving sets for graph operations were studied in (Susilowati et al. 2019).

  • Metric resolving complements for classical graphs such as complete, cycle, star, and path were examined in (Susilowati et al. 2021).

  • Complement metric dimensions for comb products of path, star, and complete graphs were explored in (Rosyidah et al., 2021).

  • Tree-like graph spaces such as caterpillars, firecracker graphs, and banana trees have been analyzed in (Amalia et al. 2021).

The present study focuses on six structured symmetric convex planar (SCP) spaces. Metric generators of the first three, Dn , Rn , and Qn , were analyzed by Imran et al. (2012), while the latter three, Sn , Tn , and Un , were examined by Imran et al. (2010). Here, we investigate the maximum cardinality subsets in each SCP space that serve as metric resolving complements. These sets indicate the worst-case configurations of landmarks that hinder navigation. Consequently, we establish the exact values of the complement metric dimension for all six SCP spaces.

2. Complement Metric Dimension of Dn

For n3, the symmetric convex planar (SCP) space Dn consists of four distinct types of points, labeled as ai , bi , ci , and di for each i[n], following the structure described in Imran et al. (2012). These vertices are respectively referred to as inner points (I), inner-attached points (IA), outer-attached points (OA), and outer points (O). Fig. 1 depicts the general half-structure of Dn , highlighting its symmetric and convex configuration.

General structural half view of the SCP space Dn .
Fig. 1.
General structural half view of the SCP space Dn .

We begin by presenting two preliminary claims that establish bounds on pairwise distances in metric resolving complements of Dn .

Claim 1 If n4 is even, then no pair (u,v) of distinct I-, IA-, OA-, or O-points contributes to a metric resolving complement of Dn whenever m(u,v)<diam(Dn)2.

Proof. It is known from Imran et al. (2012) that a minimum metric generator of Dn consists of three vertices. If all points in such a generator are selected from the same type either all I, IA, OA, or O then their resolving capability is preserved only if the inequality m x,y diam Dn 2 holds for every distinct pair x,y in the set.

In order to complete the parity-based analysis, we now present the corresponding situation when n is odd with the proof analogous to Claim 1.

Claim 2 If n5 is odd, then no pair u,v of distinct I-, IA-, OA-, or O-points contributes to a metric resolving complement of Dn whenever m u,v <diam Dn 3.

Based on these structural bounds, we now determine the exact complement metric dimension of Dn .

Theorem 3 For n4, the complement metric dimension of the SCP space Dn is given by:

dim ¯ Dn ={ 2n4, if n is even, 2n6, if n is odd.

Proof. We divide the proof into two cases based on the parity of n.

Case I: n is even. Define a set

S={ai,bj,ck,dl  ;  1in2,   3jn2,   n2kn3,   n2l<n}

containing 2n4 vertices. Then the pair bn,cn in V Dn \S satisfies cS bn =cS cn , so S is a metric resolving complement. Thus, dim ¯ Dn 2n4.

Assume, for contradiction, that a resolving complement S' exists with S' >2n4, say S' =2n3. Then there must be a pair of points in S' from the same point-type with distance less than diam Dn 2, contradicting Claim 1. Hence, dim ¯ Dn 2n4.

Having completed the analysis for even n, we now examine the complementary case when n is odd.

Case II: n is odd. Define a set

S= ai,bj,ck,dl   ;   1i n1 2,   3j n1 2,   n1 2<kn3,   n2l<n

containing 2n6 vertices. Then the pair bn,cn in V Dn \S satisfies cS bn =cS cn , so S is a metric resolving complement. Thus, dim ¯ Dn 2n6.

Assume, for contradiction, that a resolving complement S' exists with S' >2n6, say S' =2n5. Then a pair of points in S' must exist from the same point-type with distance less than diam Dn 3, contradicting Claim 2. Therefore, dim ¯ Dn 2n6.

3. Complement Metric Dimension of R n

For n3, the SCP space Rn consists of three distinct types of points labeled as ai , bi , and ci for in , as described in (Imran et al., 2012). These points correspond to inner points (I), middle points (M), and outer points (O) respectively. The structural pattern of Rn is shown in Fig. 2.

General structural half view of the SCP space Rn.
Fig. 2.
General structural half view of the SCP space Rn.

To determine the complement metric dimension of Rn , we first present two preparatory claims about intra-type distances.

Claim 4 If n6 is even, then no pair u,v of distinct I-, M-, or O-points contributes to a metric resolving complement of Rn whenever m u,v <diam Rn 1.

Proof. According to Imran et al. (2012), the minimum metric generator of Rn consists of three vertices. When the metric generator includes vertices of only one type either all I, M, or O then it can distinguish all other points only if the pairwise distances satisfy m x,y diam Rn 1 for all distinct x,yS.

A parallel restriction appears in the odd case of n, as described next.

Claim 5 If n5 is odd, then no pair u,v of distinct I-, M-, or O-points contributes to a metric resolving complement of Rn whenever m u,v <diam Rn 2.

We now derive the exact expression for the complement metric dimension of Rn .

Theorem 6 For n5, the complement metric dimension of the SCP space Rn is given by:

dim ¯ Rn ={ 3n 2, if n is even, n1, if n is odd.

Proof. We analyze both cases based on the parity of n.

Case I: n is even. Consider the set

S={ai,bj,ck   ;   1in2,   n2<j,kn}

which contains 3n 2 points. The vertices an/2+1 and bn/2 in V(Rn)\S satisfy cS an/2+1 =cS bn/2 , making S a resolving complement. Therefore, dim ¯ Rn 3n 2 .

Assume, for contradiction, that a complement exists with more than 3n 2 points. Let S' = 3n 2+1. Then some pair within the same type must violate Claim 1, i.e., m u,v <diam Rn 1, contradicting the resolving condition. Thus, dim ¯ Rn 3n 2 .

To complete the proof, we now consider the remaining case corresponding to odd n.

Case II: n is odd. Let

S={ai,bj,ck   ;   1i n1 2,   n1 2<j,kn1}

be a set of n1 points. The vertices a n+1 /2 and b n1 /2 in V Rn \S satisfy cS a n+1 /2 =cS b n1 /2 , hence S is a metric resolving complement. So, dim ¯ Rn n1.

Assume a complement S' exists with size n. Then some pair within S' must violate Claim 2, i.e., m u,v <diam Rn 2, contradicting the resolving condition. Therefore, dim ¯(Rn)n1.

4. Complement Metric Dimension of Q n

For n3, the SCP space Qn is composed of four distinct types of points, labeled ai , bi , ci , and di for each i[n], as introduced in Imran et al. (2012). These points are categorized as inner points (I), middle points (M), interior points (INT), and outer points (O), respectively. The half-structure of Qn is illustrated in Fig. 3.

General structural half view of the SCP space Qn .
Fig. 3.
General structural half view of the SCP space Qn .

We begin by stating two conditional claims related to intra-type distances in resolving complements.

Claim 7 If n6 is even, then no pair u,v of distinct I-, M-, INT-, or O-points contributes to a metric resolving complement of Qn whenever m u,v <diam Qn 2.

Proof. As established in Imran et al. (2012), three vertices suffice to form a minimum metric generator of Qn . If all selected vertices belong to the same type, they can resolve the space only if the condition m x,y diam Qn 2 holds for every pair of distinct vertices in the set.

An additional refinement is required when n is odd, as stated in the following claim with the proof similar to the Claim 7.

Claim 8 If n7 is odd, then

  • 1.

    No pair of distinct I-, M-, or O-points contributes to a metric resolving complement of Qn whenever m u,v <diam Qn 3.

  • 2.

    No pair of distinct INT-points contributes whenever m u,v <diam Qn 2.

Now, we present the exact complement metric dimension of Qn .

Theorem 9 For n6, the complement metric dimension of the SCP space Qn is given by:

dim ¯ Qn = 2n1, if n is even, 2n2, if n is odd.

Proof. We examine both cases according to the parity of n.

Case I: n is even. Define the set

S= ai,bj,ck,dl   ;   1i,jn2,   2kn2,   n2<ln

which contains 2n1 vertices. The pair bn,c1 in V Qn \S satisfies cS bn =cS c1 , implying that S is a resolving complement. Therefore, dim ¯ Qn 2n1.

Assume a larger complement of size 2n exists. Then at least one pair of same-type points in S must satisfy m u,v <diam Qn 2, violating Claim 1. Hence, dim ¯ Qn 2n1.

We now turn to the second case, which arises when n is an odd integer.

Case II: n is odd. Let

S={ai,bj,ck,dl   ;   1i,j n1 2,   2k n+1 2,   n+1 2<ln}

with S=2n2. Again, bn,c1 belong to V Qn \S and have identical codes, confirming that S is a resolving complement. So dim ¯ Qn 2n2.

Assume S' =2n1. Then either:

a pair of I-, M-, or O-points satisfies m u,v <diam Qn 3, or

a pair of INT-points satisfies m u,v <diam Qn 2,

contradicting Claim 2. Therefore, dim ¯ Qn 2n2.

5. Complement Metric Dimension of S n

For n3, the SCP space Sn comprises four types of points, denoted as ai , bi , ci , and di for in , as discussed in Imran et al. (2010). These represent inner points (I), interior points (INT), exterior points (EXT), and outer points (O) respectively. The structural arrangement is illustrated in Fig. 4.

General structural half view of the SCP space Sn .
Fig. 4.
General structural half view of the SCP space Sn .

We begin by establishing two key claims that restrict the behavior of intra-type vertex pairs in resolving complements.

Claim 10 If n6 is even, then no pair u,v of distinct I-, INT-, EXT-, or O-points contributes to a metric resolving complement of Sn whenever m u,v <diam Sn 2.

Proof. According to Imran et al. (2010), three vertices suffice to generate a minimum metric basis of Sn . When such a basis consists solely of vertices from a single type, the condition m x,y diam Sn 2 must hold for all distinct x,y in the set for it to function as a valid generator.

The odd-valued case of n follows a similar pattern, as detailed in the next claim.

Claim 11 If n7 is odd, then no pair u,v of distinct I-, INT-, EXT-, or O-points contributes to a metric resolving complement of Sn whenever m u,v <diam Sn 3.

We now state and prove the main result for this section.

Theorem 12 For n6, the complement metric dimension of the SCP space Sn is given by:

dim ¯ Sn = 2n, if n is even, 2n2, if n is odd.

Proof. We examine the two cases based on the parity of n.

Case I: n is even. Define the set

S={ai,bj,ck,dl   ;   1in2,   n2<j,k,ln}

which contains 2n vertices. The pair an/2+1 ,bn/2 in V Sn \S satisfies cS an/2+1 =cS bn/2 , showing that S is a metric resolving complement. Thus, dim ¯ Sn 2n.

Suppose a larger resolving complement S' exists with S' =2n+1. Then some pair within S' of the same type must violate Claim 1, i.e., m x,y <diam Sn 2. This contradiction implies dim ¯ Sn 2n.

For completeness, we next analyze the odd-valued case of n.

Case II: n is odd. Let

S={ai,bj,ck,dl   ;   1i n1 2,   n1 2<j,k,ln1}

with S=2n2. The pair a n+1 /2 ,b n1 /2 in V Sn \S satisfies cS a n+1 /2 =cS b n1 /2 , confirming S as a metric resolving complement. Hence, dim ¯ Sn 2n2.

Assume S' =2n1. Then there exists a pair x,y in S' of the same type such that m x,y <diam Sn 3, violating Claim 2. Thus, dim ¯ Sn 2n2.

6. Complement Metric Dimension of T n

For n3, the SCP space Tn consists of four distinct types of points, denoted by ai , bi , ci , and di for each in , as discussed in Imran et al. (2010). These are referred to as inner points (I), interior points (INT), exterior points (EXT), and outer points (O). Fig. 5 illustrates the half-structure of Tn , revealing its symmetric and layered design.

General structural half view of the SCP space Tn .
Fig. 5.
General structural half view of the SCP space Tn .

We begin with two structural claims that guide the determination of the complement metric dimension.

Claim 13 If n6 is even, then

  • 1.

    No pair of distinct I-, INT-, or EXT-points contributes to a resolving complement of Tn whenever m u,v <diam Tn 2,

  • 2.

    No pair of distinct O-points contributes whenever m u,v <diam Tn 3.

Proof. It is known from Imran et al. (2010) that three vertices form a minimum metric generator of Tn . If the generator contains vertices of the same type, its resolving property holds only when the stated bounds on distances are respected.

A comparable restriction holds when n is odd, as described in the subsequent claim.

Claim 14 If n7 is odd, then no pair of distinct I-, INT-, EXT-, or O-points contributes to a metric resolving complement of Tn whenever m u,v <diam Tn 2.

We now determine the complement metric dimension of Tn .

Theorem 15 For n6, the complement metric dimension of the SCP space Tn is given by:

dim ¯ Tn ={ 2n1, if n is even, 2n2, if n is odd.

Proof. We proceed by parity-based case analysis.

Case I: n is even. Define the set

S= ai,bj,ck,dl   ;   1in2,   n2j,kn1,   n2ln1

with cardinality 2n1. The pair an,bn in V Tn \S satisfies cS an =cS bn , verifying that S is a resolving complement. Thus, dim ¯ Tn 2n1.

Assume S' =2n. Then some pair of same-type points must violate Claim 1 either with m x,y <diam Tn 2 for non-O types, or m x,y <diam Tn 3 for O-points. This contradiction implies dim ¯ Tn 2n1.

We now analyze the second scenario, namely when n takes an odd value.

Case II: n is odd. Let

S={ai,bj,ck,dl   ;   1i n1 2,   n1 2<j,kn1,   n1 2<l<n1}

with S=2n2. The pair an,bn satisfies cS an =cS bn , establishing that S is a resolving complement. So, dim ¯ Tn 2n2.

Assume S' =2n1. Then some pair in S' violates Claim 2 by having a distance less than diam Tn 2, contradicting the resolving condition. Hence, dim ¯ Tn 2n2.

7. Complement Metric Dimension of U n

For n3, the SCP space Un comprises five distinct types of points, labeled ai , bi , ci , di , and ei for each in , as studied in Imran et al. (2010). These correspond to inner points (I), interior points (INT), two layers of exterior points (EXT), and outer points (O), respectively. The half-structure of Un is illustrated in Fig. 6.

General structural half view of the SCP space Un .
Fig. 6.
General structural half view of the SCP space Un .

The following claims provide bounds on distances among intra-type pairs in resolving complements.

Claim 16 If n6 is even, then no pair u,v of distinct I-, INT-, EXT-, or O-points contributes to a resolving complement of Un whenever m u,v <diam Un 3.

Proof. As shown in Imran et al. (2010), three vertices form a minimum resolving set for Un . If all selected points are of the same type, their distinguishing power is preserved only if their pairwise distances meet or exceed diam Un 3.

The corresponding condition for odd values of n is stated in the next claim.

Claim 17 If n7 is odd, then no pair u,v of distinct I-, INT-, EXT-, or O-points contributes to a metric resolving complement of Un whenever m u,v <diam Un 4.

We now state and prove the final main result.

Theorem 18 For n6, the complement metric dimension of the SCP space Un is given by:

dim ¯ Un = 5n8 2, if n is even, 5n13 2, if n is odd.

Proof. We consider both parity cases for n.

Case I: n is even. Define the set

S= ai,bj,ck,dl,em   ;   1i,jn2,   3kn2, n2<ln3,   n2<m<n

with S= 5n8 2 . The pair cn,dn in V Un \S satisfies cS cn =cS dn , showing that S is a resolving complement. Hence, dim ¯ Un 5n8 2 .

Suppose a larger resolving complement S' exists with S' = 5n8 2+1. Then a pair of same-type points in S' must violate Claim 1 by having a distance less than diam Un 3, contradicting the resolving property. Thus, dim ¯ Un 5n8 2 .

Having resolved the even case, we proceed to investigate the odd case of n.

Case II: n is odd. Let

S= ai,bj,ck,dl,em   ;   1i,j n1 2,   3k n1 2,   n1 2<ln3,   n1 2<m<n

with S= 5n13 2 . The pair cn,dn in V Un \S satisfies cS cn =cS dn , proving that S is a resolving complement. Therefore, dim ¯ Un 5n13 2 .

Assume a complement S' of size 5n13 2+1 exists. Then some same-type pair in S' must satisfy m u,v <diam Un 4, violating Claim 2. Hence, dim ¯ Un 5n13 2 .

Table 1 summarizes the final results obtained from each SCP space analyzed:

Table 1. Complement metric dimensions of the six SCP spaces.
SCP space Order of SCP CMD for even n CMD for Odd n
Dn 4n 2n4 2n6
Rn 3n 3n 2 n1
Qn 4n 2n1 2n2
Sn 4n 2n 2n2
Tn 4n 2n1 2n2
Un 5n 5n8 2 5n13 2

8. Concluding Remarks and Summary

This study explored the structural limitations in landmark-based navigation through the lens of complement metric dimension (CMD) in six symmetric convex planar (SCP) spaces. Each SCP space was analyzed to determine the maximum cardinality of a metric resolving complement a set of points that fails to distinguish all other points based on metric codes.

These complements model the concept of navigational ambiguity in robot pathfinding. Unlike traditional metric generators which aid in unique identification, resolving complements represent configurations that cause maximum confusion in spatial navigation. By characterizing these worst-case sets, our work contributes to understanding robustness limits in graph-based localization systems.

The exact values of the complement metric dimensions were derived through constructive combinatorial arguments and upper-bounding claims based on structural diameter restrictions. Table 1 summarizes the final results obtained for each SCP space analyzed:

Due to the symmetric and layered nature of these SCP spaces, we observe a consistent trend: the complement metric dimension tends to exceed half of the total number of vertices (order) in the graph. This structural insight motivates the following conjecture for future investigation.

Conjecture: The complement metric dimension of any SCP space X with order VX 3 satisfies

dim ¯X> VX /2.

This work lays the groundwork for further explorations into generalized classes of convex polytopes, higher-dimensional grid spaces, and applications in uncertainty modeling for robotics, sensor networks, and chemical graph theory.

CRediT authorship contribution statement

Muhammad Salman: Conceptualization, methodology and supervision; Arslan Iqbal: Investigation and writing original draft preparation; Muhammad Imran: Formal analysis and funding acquisition. All authors have read and agreed to this version of the manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Declaration of generative AI and AI-assisted technologies in the writing process

The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.

Funding

This research is self-funded.

References

  1. , , , , . The complement metric dimension of particular tree. J Phys: Conf Ser. 2021;1836:012011. https://doi.org/10.1088/1742-6596/1836/1/012011
    [Google Scholar]
  2. , . Introduction to graphs. In: Chromatic graph theory Chromatic graph theory. Chapman and Hall/CRC; . p. :27-52. https://doi.org/10.1201/9780429438868-2
    [Google Scholar]
  3. , , , . Resolvability in graphs and the metric dimension of a graph. Discrete Appl Math. 2000;105:99-113. https://doi.org/10.1016/s0166-218x(00)00198-0
    [Google Scholar]
  4. , . On the metric dimenson of a graph. Ars Combinatoria. 1976;2:191-195. https://combinatorialpress.com/ars/vol2/
    [Google Scholar]
  5. , , . Families of plan graphs with constant metric dimension. Utilitas Mathematica. 2012;88:43-57. https://combinatorialpress.com/um/vol88/
    [Google Scholar]
  6. , , . On families of convex polytopes with constant metric dimension. Comput Math Appl. 2010;60:2629-2638. https://doi.org/10.1016/j.camwa.2010.08.090
    [Google Scholar]
  7. , , . Landmarks in graphs. Discrete Appl Math. 1996;70:217-229. https://doi.org/10.1016/0166-218x(95)00106-2
    [Google Scholar]
  8. Rosyidah, N.M., Zahidah, S., Purwati, U.D., Susilowati, L., 2021. On comb product graphs with respect to the complement metric dimension. International conference on mathematics, computational sciences and statistics 2020 Surabaya, Indonesia, pp. 020006. https://doi.org/10.1063/5.0042618
  9. , , . Resolving share and topological index. Utilitas Mathematica. 2018;108:89-105. https://combinatorialpress.com/um/vol108/
    [Google Scholar]
  10. . Leaves of trees. Congressus Numerantium. 1975;14:549-559.
    [Google Scholar]
  11. Susilowati, L., Nurrona, A., Purwati, U.D., 2021. The complement metric dimension of the joint graph. 4TH International sciences, technology and engineering conference (ISTEC) 2020: Exploring Materials for the Future Arau, Malaysia, pp. 020003. https://doi.org/10.1063/5.0042149
  12. , , . The complement metric dimension of graphs and its operations. Int J Civil Eng Technol. 2019;10:2386-2396. https://iaeme.com/Home/article_id/IJCIET_10_03_239
    [Google Scholar]
Show Sections