Loading [MathJax]/jax/element/mml/optable/GeneralPunctuation.js
Research article Special Issues

Common fixed point of nonlinear contractive mappings

  • The purpose of this paper is to study the existence of a common fixed point for a pair of mappings without assumption of the contractive coefficient being fixed and less than 1. By replacing the fixed contractive coefficient with a nonlinear contractive function, we establish a unique common fixed point theorem for a pair of asymptotically regular self-mappings with either orbital continuity or q-continuity in a metric space. Moreover, by the asymptotical regularity of two approximate mappings, we prove that a pair of nonexpansive and continuous self-mappings, which are defined on a nonempty closed convex subset of a Banach space, have a common fixed point. Some examples are given to illustrate that our results are extensions of a recent result in the existing literature.

    Citation: Hui Huang, Xue Qian. Common fixed point of nonlinear contractive mappings[J]. AIMS Mathematics, 2023, 8(1): 607-621. doi: 10.3934/math.2023028

    Related Papers:

    [1] Saif Ur Rehman, Iqra Shamas, Shamoona Jabeen, Hassen Aydi, Manuel De La Sen . A novel approach of multi-valued contraction results on cone metric spaces with an application. AIMS Mathematics, 2023, 8(5): 12540-12558. doi: 10.3934/math.2023630
    [2] Dong Ji, Yao Yu, Chaobo Li . Fixed point and endpoint theorems of multivalued mappings in convex b-metric spaces with an application. AIMS Mathematics, 2024, 9(3): 7589-7609. doi: 10.3934/math.2024368
    [3] Yan Han, Shaoyuan Xu, Jin Chen, Huijuan Yang . Fixed point theorems for b-generalized contractive mappings with weak continuity conditions. AIMS Mathematics, 2024, 9(6): 15024-15039. doi: 10.3934/math.2024728
    [4] Shaoyuan Xu, Yan Han, Suzana Aleksić, Stojan Radenović . Fixed point results for nonlinear contractions of Perov type in abstract metric spaces with applications. AIMS Mathematics, 2022, 7(8): 14895-14921. doi: 10.3934/math.2022817
    [5] Mohamed Gamal, Tahair Rasham, Watcharaporn Cholamjiak, Fu-Gui Shi, Choonkil Park . New iterative scheme for fixed point results of weakly compatible maps in multiplicative GMmetric space via various contractions with application. AIMS Mathematics, 2022, 7(8): 13681-13703. doi: 10.3934/math.2022754
    [6] Abdullah Shoaib, Tahair Rasham, Giuseppe Marino, Jung Rye Lee, Choonkil Park . Fixed point results for dominated mappings in rectangular b-metric spaces with applications. AIMS Mathematics, 2020, 5(5): 5221-5229. doi: 10.3934/math.2020335
    [7] Pragati Gautam, Vishnu Narayan Mishra, Rifaqat Ali, Swapnil Verma . Interpolative Chatterjea and cyclic Chatterjea contraction on quasi-partial b-metric space. AIMS Mathematics, 2021, 6(2): 1727-1742. doi: 10.3934/math.2021103
    [8] Qing Yang, Chuanzhi Bai . Fixed point theorem for orthogonal contraction of Hardy-Rogers-type mapping on O-complete metric spaces. AIMS Mathematics, 2020, 5(6): 5734-5742. doi: 10.3934/math.2020368
    [9] Muhammad Nazam, Aftab Hussain, Asim Asiri . On a common fixed point theorem in vector-valued b-metric spaces: Its consequences and application. AIMS Mathematics, 2023, 8(11): 26021-26044. doi: 10.3934/math.20231326
    [10] Mohammed Shehu Shagari, Akbar Azam . Integral type contractions of soft set-valued maps with application to neutral differential equations. AIMS Mathematics, 2020, 5(1): 342-358. doi: 10.3934/math.2020023
  • The purpose of this paper is to study the existence of a common fixed point for a pair of mappings without assumption of the contractive coefficient being fixed and less than 1. By replacing the fixed contractive coefficient with a nonlinear contractive function, we establish a unique common fixed point theorem for a pair of asymptotically regular self-mappings with either orbital continuity or q-continuity in a metric space. Moreover, by the asymptotical regularity of two approximate mappings, we prove that a pair of nonexpansive and continuous self-mappings, which are defined on a nonempty closed convex subset of a Banach space, have a common fixed point. Some examples are given to illustrate that our results are extensions of a recent result in the existing literature.



    Fixed point theory is an powerful tool of modern mathematics and may be viewed as a core context of nonlinear analysis. Since this theory has many applications in computer science, engineering, physics, economics, optimization theory, etc, a number of excellent books on it have been published. The reader is referred to the books [1,6,10] for more details on fixed point theory and applications. With the development of science, in recent years, the idea and techniques of fixed point theory have been used to study the problem of finite-/fixed-time synchronization for Clifford-valued recurrent neural networks with time-varying delays [4] and investigate control of a symmetric chaotic supply chain system [17].

    Probably the most well known result in fixed point theory is Banach's contraction mapping principle. For the convenience of describing it, we need some notations. Let (X,d) be a metric space, F:XX be a mapping and λ[0,1] with

    d(Fx,Fy)λd(x,y),   for all x,yX. (1.1)

    If λ<1 in (1.1), the mapping F is called a contractive mapping, and λ is called a contractive coefficient of F, whereas if λ=1, F is called a nonexpansive mapping. Let R denote the set of all real numbers and N denote the set of all positive integer numbers. For simplification, for any xX and nN, we define Fnx inductively by F1x=Fx and Fn+1x=F(Fnx). Moreover, we define F0x=x.

    Banach's contraction mapping principle is stated as follows.

    Theorem 1.1. Let (X,d) be a complete metric space and F:XX be a contractive mapping. Then F has a unique fixed point uX. Furthermore, for any xX, the sequence {Fnx} converges to u.

    In order to extend Banach's contraction mapping principle, Geraghty [8] introduced a class of test functions.

    Definition 1.2. [8] Let R+:={tR | t0}. S is defined as the set of functions ψ:R+[0,1] with the properties:

    (ⅰ) 0ψ(t)<1, for all t>0,

    (ⅱ) limnψ(tn)=1 implies limntn=0.

    Then using Definition 1.2, Geraghty [8] established the following result.

    Theorem 1.3. Let (X,d) be a complete metric space, F:XX be a mapping and ψS. If

    d(Fx,Fy)ψ(d(x,y))d(x,y),   for all x,yX,

    then F has a unique fixed point uX.

    One can deduce that the function F in Theorems 1.1 and 1.3 is a uniformly continuous function. In order to obtain a fixed point under weaker condition than Theorems 1.1 and 1.3, some useful concepts are given by some researchers. Browder and Petryshyn [5] introduced the concept of asymptotical regularity as follows.

    Definition 1.4. Let (X,d) be a metric space, F:XX be a mapping. F is said to be asymptotically regular on X, if

    limnd(Fnx,Fn+1x)=0,   for all xX.

    ˊCiriˊc [7] introduced the definition of orbital continuity as follows.

    Definition 1.5. [7] Let (X,d) be a metric space, F:XX be a mapping. The set O(F,z):={Fnz | n=0,1,2,} is called the orbit of F at the point zX. F is said to be orbitally continuous at yX if for any sequence {xk}O(F,x) for some xX, limkxk=y implies that limkFxk=Fy. We say that F is orbitally continuous on X if F is orbitally continuous at each point yX.

    It is known that every continuous self-mapping is orbitally continuous, but the converse is not true [7].

    Pant and Pant [15] introduced the definition of q-continuity for a qN as follows.

    Definition 1.6. [15] Let (X,d) be a metric space, F:XX be a mapping and qN. F is said to be q-continuous at yX, if for each sequence {xn}X, limnFq1xn=y implies that limnFqxn=Fy. We say that F is q-continuous on X if F is q-continuous at each point yX.

    Clearly, 1-continuity is the same as continuity. It is known that q-continuity implies that q+1-continuity, but the converse is not true [15].

    Recently, Bisht [2] weakened the conditions of Gˊornicki's [9] result from continuity to q-continuity or orbital continuity, and obtained the following result.

    Theorem 1.7. Let (X,d) be a complete metric space, F:XX be a mapping, qN, θ[0,1) and τ0. Suppose that F is asymptotically regular, q-continuous or orbitally continuous on X, and satisfy

    d(Fx,Fy)θd(x,y)+τ(d(x,Fx)+d(y,Fy)),   for all x,yX.

    Then F has a unique fixed point uX and limnFnx=u.

    One can observe that a fixed point of F in fact is a coincidence point of F with the identity mapping I on X. Therefore, replacing the identity mapping I by another self-mapping G of X to obtain a common fixed point of F and G is natural. In 2020, Bisht and Singh [3] extended Theorem 1.7 from a single self-mapping F to a pair of mappings F and G, and obtained the following theorem.

    Theorem 1.8. Let (X,d) be a complete metric space, F, G:XX be two mappings, θ[0,1) and τ0. Suppose that F is asymptotically regular with respect to G, and F and G are (F,G)-orbitally continuous and compatible. Further, F and G satisfy

    d(Fx,Fy)θd(Gx,Gy)+τ(d(Fx,Gx)+d(Fy,Gy)),   for all x,yX.

    Then F and G have a unique common fixed point uX.

    In 2020, unlike conditions given in Theorem 1.8, Khan and Oyetunbi [14] also extended Theorem 1.7 to a pair of self-mapping F and G, and obtained the following result.

    Theorem 1.9. Let (X,d) be a complete metric space, F, G:XX be two mappings, qN, θ[0,1) and τ0. Suppose that F and G are asymptotically regular, q-continuous or orbitally continuous on X, and satisfy

    d(Fx,Gy)θd(x,y)+τ(d(x,Fx)+d(y,Gy)),   for all x,yX. (1.2)

    Then F and G have a unique common fixed point uX and limnFnx=limnGnx=u.

    Very recently, the study of fixed point theory has received a growing interest and made great progress. In 2020, Hassan et al. [12] introduced S-iteration scheme for approximation of fixed point of the nonexpansive mappings and proved that it is stable and faster than some iteration schemes existing in the literature. They also established some convergence theorems for Suzuki's generalized nonexpansive mappings in uniformly convex Banach spaces. In 2021, Hammad, Agarwal and Guirao [11] presented some tripled fixed point results for a pair of mappings under (φ,ρ,ι)-contraction in ordered partially metric spaces. As applications, they also discussed the existence and uniqueness of the solution to an initial value problem and a homotopy theory. In 2022, Rasham et al. [16] established some theorems on common fixed points of set-valued mappings under αψ ˊCiriˊc type contraction in a complete modular like metric space. They also proved that a pair of multi-graph dominated mappings with graph contractions have a common fixed point.

    In Theorem 1.9, we observe that the contractive coefficient θ is fixed and less than 1. In Theorem 1.3, we also observe that the contractive coefficient θ in Theorem 1.1 can be replaced by a function ψS. In this paper, motivated by the above observations, we extend Theorem 1.9 for a pair of self-mappings F and G by replacing θ in (1.2) with a function ψS or 1 in order to get common fixed points for more mathematical models. Firstly, by replacing θ with a function ψS, we obtain that asymptotically regular, either orbitally continuous or q-continuous mappings F and G have a unique common fixed point in a complete metric space. Secondly, by replacing θ with 1 and modifying conditions of Theorem 1.9, we also obtain that F and G have a common fixed point on a nonempty, closed convex subset of a Banach space. Some examples are given to illustrate our extensions.

    The following lemma plays an important role in the proof of Theorems 2.2 and 2.5.

    Lemma 2.1. [18] Let {un}, {vn} and {wn} be three real sequences with un0 and vn(0,1). Suppose that

    (ⅰ) un+1(1vn)un+vnwn;

    (ⅱ) n=1vn=;

    (ⅲ) lim supnwn0 or n=1vnwn is convergent.

    Then limnun=0.

    Now we establish our one main result.

    Theorem 2.2. Let (X,d) be a complete metric space, and F and G be two asymptotically regular self-mappings on X. Suppose that there exist a function ψS and a constant τ[0,+) such that

    d(Fx,Gy)ψ(d(x,y))d(x,y)+τ(d(x,Fx)+d(y,Gy)),   for all x,yX, (2.1)

    and that F and G are either orbitally continuous or q-continuous for some qN on X. Then F and G have a unique common fixed point on X.

    proof. Take any xX. Define xn=Fnx and yn=Gnx for all nN. By (2.1), we have

        d(Fn+1x,Gn+1x)=d(F(Fnx),G(Gnx))ψ(d(Fnx,Gnx))d(Fnx,Gnx)+τ(d(Fnx,Fn+1x)+d(Gnx,Gn+1x)). (2.2)

    Since 0ψ(d(Fnx,Gnx))1, we can divide lim supnψ(d(Fnx,Gnx)) into two cases.

    Case Ⅰ: lim supnψ(d(Fnx,Gnx))=1.

    In this case, there exists a subsequence {ψ(d(Fnkx,Gnkx))} of {ψ(d(Fnx,Gnx))} such that

    limkψ(d(Fnkx,Gnkx))=1. (2.3)

    Since ψ is a function satisfying Definition 1.2, with (2.3), we conclude

    limkd(Fnkx,Gnkx)=0. (2.4)

    In the following, we will prove that {Fnkx} is a Cauchy sequence. Suppose to the contrary that {Fnkx} is not a Cauchy sequence, then there exist a ε0>0 and two integer number sequences {n˜k(i)}, {nk(i)} of {nk} with n˜k(i)>nk(i)>i such that

    d(Fn˜k(i)x,Fnk(i)x)ε0,   i=1,2,. (2.5)

    Consequently, we have

    ε0d(Fn˜k(i)x,Fnk(i)x)d(Fn˜k(i)x,Fn˜k(i)1x)+d(Fn˜k(i)1x,Fnk(i)1x)+d(Fnk(i)1x,Fnk(i)x). (2.6)

    Letting i in (2.6), by the asymptotical regularity of F, we obtain

    lim infid(Fn˜k(i)1x,Fnk(i)1x)ε0. (2.7)

    Now, by (2.1), we have

    d(Fn˜k(i)x,Fnk(i)x)d(Fn˜k(i)x,Gnk(i)x)+d(Gnk(i)x,Fnk(i)x)ψ(d(Fn˜k(i)1x,Gnk(i)1x))d(Fn˜k(i)1x,Gnk(i)1x)+τ(d(Fn˜k(i)1x,Fn˜k(i)x)+d(Gnk(i)1x,Gnk(i)x))+d(Gnk(i)x,Fnk(i)x)ψ(d(Fn˜k(i)1x,Gnk(i)1x))(d(Fn˜k(i)1x,Fn˜k(i)x)+d(Fn˜k(i)x,Fnk(i)x)+d(Fnk(i)x,Gnk(i)x)+d(Gnk(i)x,Gnk(i)1x))+τ(d(Fn˜k(i)1x,Fn˜k(i)x)+d(Gnk(i)1x,Gnk(i)x))+d(Gnk(i)x,Fnk(i)x). (2.8)

    Dividing both sides of inequality (2.8) by d(Fn˜k(i)x,Fnk(i)x), with (2.5), we conclude

    1ψ(d(Fn˜k(i)1x,Gnk(i)1x))(d(Fn˜k(i)1x,Fn˜k(i)x)d(Fn˜k(i)x,Fnk(i)x)+1+d(Fnk(i)x,Gnk(i)x)d(Fn˜k(i)x,Fnk(i)x)+d(Gnk(i)x,Gnk(i)1x)d(Fn˜k(i)x,Fnk(i)x))+τd(Fn˜k(i)1x,Fn˜k(i)x)+d(Gnk(i)1x,Gnk(i)x)d(Fn˜k(i)x,Fnk(i)x)+d(Gnk(i)x,Fnk(i)x)d(Fn˜k(i)x,Fnk(i)x). (2.9)

    Letting i in (2.9), by (2.4), (2.5), asymptotical regularity of F and G, and the fact 0ψ()1, we deduce that

    limiψ(d(Fn˜k(i)1x,Gnk(i)1x))=1.

    Since ψ satisfies Definition 1.2, we conclude

    limid(Fn˜k(i)1x,Gnk(i)1x)=0. (2.10)

    Since

    d(Fn˜k(i)1x,Fnk(i)1x)d(Fn˜k(i)1x,Gnk(i)1x)+d(Gnk(i)1x,Fnk(i)1x),

    by combining (2.4) and (2.10), we conclude

    limid(Fn˜k(i)1x,Fnk(i)1x)=0,

    which contradicts (2.7). Therefore, {xnk}={Fnkx} is a Cauchy sequence. Since X is complete, {xnk} converges to a point u in X. Since

    d(Gnkx,u)d(Gnkx,Fnkx)+d(Fnkx,u),

    with (2.4), we deduce that {Gnkx} also converges to u.

    Now, assume that F is orbitally continuous. Since {xnk} converges to u, the orbital continuity of F implies that {Fxnk} converges to Fu. By asymptotical regularity of F, we have

    limkd(xnk+1,xnk)=limkd(Fnk+1x,Fnkx)=0,

    resulting in

    limkFxnk=limkxnk+1=limkxnk=u.

    Uniqueness of limit implies that Fu=u.

    Now, suppose that F is q-continuous. Since

    d(xnk+j,xnk)d(xnk+j,xnk+j1)++d(xnk+1,xnk)=d(Fnk+jx,Fnk+j1x)++d(Fnk+1x,Fnkx),   j=1,2,,q,

    using the asymptotical regularity of F, we conclude that

    limkxnk+j=limkxnk=u,   j=1,2,,q, (2.11)

    especially,

    limkFq1xnk=limkxnk+q1=u. (2.12)

    Since F is q-continuous, with (2.12), we have

    limkFqxnk=Fu. (2.13)

    However, from (2.11), we have

    limkFqxnk=limkxnk+q=u. (2.14)

    Combining (2.13) and (2.14), we obtain Fu=u.

    Since {Gnkx} also converges to u, similarly, orbital continuity or q-continuity of G implies that Gu=u.

    Now, u is a common fixed point of F and G. Assume that v is another common fixed point of F and G with uv. Since d(u,v)>0, it follows from Definition 1.2 that 0ψ(d(u,v))<1. By (2.1), we have

    d(u,v)=d(Fu,Gv)ψ(d(u,v))d(u,v)+τ(d(u,Fu)+d(v,Gv))=ψ(d(u,v))d(u,v)<d(u,v),

    which leads to a contradiction. Therefore, the common fixed point of F and G is unique.

    Case Ⅱ: lim supnψ(d(Fnx,Gnx))<1.

    In this case, there exists a σ(0,1) such that 0ψ(d(Fnx,Gnx))<σ. With (2.2), we have

    d(Fn+1x,Gn+1x)σd(Fnx,Gnx)+τ(d(Fnx,Fn+1x)+d(Gnx,Gn+1x)). (2.15)

    Let un:=d(Fnx,Gnx), vn:=1σ, and

    wn:=τ(d(Fnx,Fn+1x)+d(Gnx,Gn+1x))1σ.

    From (2.15), we have

    un+1(1vn)un+vnwn,   for all nN.

    Since F and G are asymptotically regular on X, we conclude that limnwn=0. Moreover,

    n=1vn=n=1(1σ)=.

    By Lemma 2.1, we conclude

    limnd(Fnx,Gnx)=0.

    Then for any subsequence {nk} of {n}, we have

    limkd(Fnkx,Gnkx)=0,

    that is, (2.4) holds. From here, the proof is the same as previous Case I, so we omit it. The proof is completed.

    In the following, we give two examples to illustrate that Theorem 2.2 is an extension of Theorem 1.9. Or, more specifically, Theorem 2.2 can be applied to the two examples but Theorem 1.9 can not.

    Example 2.3. Let X=[0,1] be equipped with the metric d(x,y)=|xy|, and F, G:XX be defined by

    Fx=x14x2,   Gx=0,   for all xX.

    We first prove that F and G do not satisfy inequality (1.2). Suppose to the contrary that F and G satisfy inequality (1.2):

    d(Fx,Gy)θd(x,y)+τ(d(x,Fx)+d(y,Gy)),   for all x, yX. (2.16)

    Taking x=1n, y=0 in (2.16) for all nN, we have

    |1n14n2|θ1n+τ14n2,

    and so

    |114n|θ+τ14n. (2.17)

    By sending n in (2.17), we obtain 1θ, which contradicts θ[0,1). Therefore, F and G do not satisfy conditions of Theorem 1.9.

    Now, define ψ:R+[0,1] by

    ψ(t)={112t,    for all t[0,1],12,      for all t(1,+).

    Then ψS. For any x,yX, since |xy|1, we have

    ψ(d(x,y))d(x,y)+(d(x,Fx)+d(y,Gy))d(Fx,Gy)=(112|xy|)|xy|+14x2+y(x14x2)=|xy|12(xy)2+12x2+yx={12x212(xy)2,                   if xy,2(yx)12(yx)2+12x2,  if x<y,0.

    Therefore, F and G satisfy (2.1) with τ=1. Clearly, F and G are continuous on X and G is asymptotically regular on X. For any xX, since

    0Fn+1x=Fnx14(Fnx)2Fnx,

    we conclude that {Fnx} is a decreasing sequence with a lower bound 0. Hence, {Fnx} converges to a point in X. This implies that F is an asymptotically regular mapping. All conditions of Theorem 2.2 are satisfied, so F and G have a unique common fixed point. In fact, 0 is the unique common fixed point of F and G.

    Example 2.4. Let X=[1,1] be equipped with the metric d(x,y)=|xy|, and F, G:XX be defined by

    Fx={x1+x,   0x1,12,        1x<0,Gx={0,    0x1,1,    1x<0. (2.18)

    We first prove that F and G do not satisfy inequality (1.2). Suppose to the contrary that F and G satisfy

    d(Fx,Gy)θd(x,y)+τ(d(x,Fx)+d(y,Gy)),   for all x, yX. (2.19)

    Taking x=1n and y=0 in (2.19) for all nN, we obtain

    nn+1θ+τ1n+1. (2.20)

    By sending n in (2.20), we have 1θ, which contradicts θ[0,1). Therefore, F and G do not satisfy conditions of Theorem 1.9.

    Now, define ψ:R+[0,1] by

    ψ(t)=11+t,    for all t[0,+).

    Then ψS. In the following, we will prove that for any x,yX,

    Ψ:=ψ(d(x,y))d(x,y)+4(d(x,Fx)+d(y,Gy))d(Fx,Gy)0. (2.21)

    To prove inequality (2.21), we divide x and y into four cases.

    Case 1: x0 and y0. In this case,

    Ψ=|xy|1+|xy|+4x21+x+4yx1+x{|xy|1+|xy|+4x21+x+4xx,   if yx,xy1+x+4x21+x+4yx1+x,   if y<x,0.

    Case 2: x0 and y<0. In this case,

    Ψ=|xy|1+|xy|+4x21+x+4|y1||x1+x1|=|xy|1+|xy|+4x21+x+4|y|+4+x1+x10.

    Case 3: x<0 and y<0. In this case,

    Ψ=|xy|1+|xy|+4|x12|+4(|y|+1)120.

    Case 4: x<0 and y0. In this case,

    Ψ=|xy|1+|xy|+4|x12|+4|y|12=|xy|1+|xy|+4|x|+2+4|y|120.

    From the above four cases, we deduce that (2.21) holds. Therefore, F and G satisfy (2.1) with τ=4.

    Clearly, F is 2-continuous at all points in X{0} since F is continuous at these points. Now, assume that limnFxn=0. By (2.18), we have 0xn1 and 0Fxn1 when n sufficiently large. Thus,

    limnF2xn=limnFxn1+Fxn=0=F0.

    Therefore, F is 2-continuous at 0. Similarly, we can also prove that G is 2-continuous on X.

    Since for any xX,

    limnd(Fnx,Fn+1x)=limn{|x1+nxx1+(n+1)x|,   if  0x1,|11+n12+n|,               if 1x<0,=0,

    we deduce that F is asymptotically regular on X. Similarly, we can also prove that G is asymptotically regular on X.

    We have verified that F and G satisfy all conditions of Theorem 2.2, and hence, they have a unique common fixed point on X. In fact, 0 is the unique common fixed point of F and G.

    Now, we establish our another main result.

    Theorem 2.5. Let X be a Banach space, D be a nonempty closed convex subset of X, F, G:DD be two mappings, r(0,1), τ[0,+), zD and

    ˉFx:=rFx+(1r)z,    ˉGx:=rGx+(1r)z,   for all xD.

    Suppose that the following conditions hold:

    (ⅰ) ;

    (ⅱ) \bar{F} and \bar{G} are asymptotically regular on D ;

    (ⅲ) \lim\limits_{n\rightarrow \infty}\left\|\bar{F}^{n}x-F(\bar{F}^{n}x)\right\| = 0, \ \ \ \lim\limits_{n\rightarrow \infty}\left\|\bar{G}^{n}x-G(\bar{G}^{n}x)\right\| = 0, \ \ \ \hbox{for all} \ x\in D ;

    (ⅳ) F and G are continuous on D .

    Then F and G have a common fixed point on D .

    proof. Since the proof is similar with the proof Theorem 2.2, we just give a sketchy proof here. Take any x\in D . For all n\in \mathbb{N} , by condition (ⅰ), we have

    \begin{equation} \begin{aligned} &\left\|\bar{F}^{n+1}x-\bar{G}^{n+1}x\right\| = r\left\|F(\bar{F}^{n}x)-G(\bar{G}^{n}x)\right\|\\ &\ \ \ \leq r\left\|\bar{F}^{n}x-\bar{G}^{n}x\right\|+\tau r\left(\left\|\bar{F}^{n}x-F(\bar{F}^{n}x)\right\|+\left\|\bar{G}^{n}x-G(\bar{G}^{n}x)\right\|\right). \end{aligned} \end{equation} (2.22)

    Let u_{n}: = \left\|\bar{F}^{n}x-\bar{G}^{n}x\right\| , v_{n}: = 1-r , and

    w_{n}: = \frac{ \tau r \left(\left\|\bar{F}^{n}x-F(\bar{F}^{n}x)\right\|+\left\|\bar{G}^{n}x-G(\bar{G}^{n}x)\right\|\right)}{1-r}.

    With Lemma 2.1, condition (ⅲ) and (2.22), we conclude

    \begin{equation} \lim\limits_{n\rightarrow \infty}\left\|\bar{F}^{n}x-\bar{G}^{n}x\right\| = 0. \end{equation} (2.23)

    Now, we prove that \{\bar{F}^{n}x\} is a Cauchy sequence. Suppose to the contrary that \{\bar{F}^{n}x\} is not a Cauchy sequence, then there exist a \varepsilon_{0} > 0 and two integer number sequences \{\tilde{n}(i)\} , \{n(i)\} with \tilde{n}(i) > n(i) > i such that

    \begin{equation} \left\|\bar{F}^{\tilde{n}(i)}x-\bar{F}^{n(i)}x\right\|\geq \varepsilon_{0}, \ \ \ i = 1, 2, \ldots . \end{equation} (2.24)

    Now, by condition (ⅰ), we have

    \begin{array}{l} \left\|\bar{F}^{\tilde{n}(i)}x-\bar{F}^{n(i)}x\right\|\leq \left\|\bar{F}^{\tilde{n}(i)}x-\bar{G}^{n(i)}x\right\|+\left\|\bar{G}^{n(i)}x-\bar{F}^{n(i)}x\right\|\\ = r\left\|F(\bar{F}^{\tilde{n}(i)-1}x)-G(\bar{G}^{n(i)-1}x)\right\|+\left\|\bar{G}^{n(i)}x-\bar{F}^{n(i)}x\right\|\\ \leq r \left\|\bar{F}^{\tilde{n}(i)-1}x-\bar{G}^{n(i)-1}x\right\|\\ +r\tau \left(\left\|\bar{F}^{\tilde{n}(i)-1}x-F(\bar{F}^{\tilde{n}(i)-1}x)\right\|+\left\|\bar{G}^{n(i)-1}x-G(\bar{G}^{n(i)-1}x)\right\|\right)+\left\|\bar{G}^{n(i)}x-\bar{F}^{n(i)}x\right\|\\ \leq r \left(\left\|\bar{F}^{\tilde{n}(i)-1}x-\bar{F}^{\tilde{n}(i)}x\right\|+\left\|\bar{F}^{\tilde{n}(i)}x-\bar{F}^{n(i)}x\right\|+\left\|\bar{F}^{n(i)}x-\bar{G}^{n(i)}x\right\| +\left\|\bar{G}^{n(i)}x-\bar{G}^{n(i)-1}x\right\|\right)\\ +r\tau \left(\left\|\bar{F}^{\tilde{n}(i)-1}x-F(\bar{F}^{\tilde{n}(i)-1}x)\right\|+\left\|\bar{G}^{n(i)-1}x-G(\bar{G}^{n(i)-1}x)\right\|\right)+\left\|\bar{G}^{n(i)}x-\bar{F}^{n(i)}x\right\|. \end{array} (2.25)

    Dividing both sides of inequality (2.25) by \left\|\bar{F}^{\tilde{n}(i)}x-\bar{F}^{n(i)}x\right\| , and with (2.24), we have

    \begin{equation} \begin{aligned} &1 \leq r \left(\frac{\left\|\bar{F}^{\tilde{n}(i)-1}x-\bar{F}^{\tilde{n}(i)}x\right\|}{\varepsilon_{0}}+1+\frac{\left\|\bar{F}^{n(i)}x-\bar{G}^{n(i)}x\right\|}{\varepsilon_{0}} +\frac{\left\|\bar{G}^{n(i)}x-\bar{G}^{n(i)-1}x\right\|}{\varepsilon_{0}}\right)\\ &\ \ \ +r\tau \frac{\left\|\bar{F}^{\tilde{n}(i)-1}x-F(\bar{F}^{\tilde{n}(i)-1}x)\right\|+\left\|\bar{G}^{n(i)-1}x-G(\bar{G}^{n(i)-1}x)\right\|}{\varepsilon_{0}}+\frac{\left\|\bar{G}^{n(i)}x-\bar{F}^{n(i)}x\right\|}{\varepsilon_{0}}. \end{aligned} \end{equation} (2.26)

    Letting i\rightarrow \infty in (2.26), by using (2.23), conditions (ⅱ) and (ⅲ), we deduce that 1\leq r . This is a contradiction since r < 1 . Consequently, \{\bar{F}^{n}x\} is a Cauchy sequence. Since D is a closed set, \{\bar{F}^{n}x\} converges to a point u in D . This and (2.23) imply that \{\bar{G}^{n}x\} also converges to u . From condition (ⅳ), F and G are continuous on D , and with condition (ⅲ) we get

    \begin{equation*} Fu = u = Gu. \end{equation*}

    Hence u is a common fixed point of F and G .

    Remark 2.6. In the following, we will give two examples to illustrate that Theorem 2.5 is an extension of Theorem 1.9, and the common fixed point of F and G which satisfy the conditions of Theorem 2.5 probably is not unique.

    Example 2.7. Let X = \mathbb{R} be equipped with the norm \|x\| = |x| , D = [-1, 1] , and F, \ G: D\rightarrow D be defined by

    \begin{equation*} Fx = x, \ \ \ \hbox{for all} \ x\in D, \end{equation*}

    and

    \begin{equation*} Gx = \left\{ \begin{aligned} &0, \ \ \ \ \ \ \ x\in [0, 1], \\ &x, \ \ \ \ \ \ \ x\in [-1, 0). \end{aligned} \right. \end{equation*}

    Obviously, the metric induced by the norm is

    d(x, y) = |x-y|, \ \ \ \forall \ x, y\in X.

    We first prove that F and G do not satisfy inequality (1.2) on D . Suppose to the contrary that F and G satisfy inequality (1.2):

    \begin{equation} d(Fx, Gy)\leq \theta d(x, y)+\tau(d(x, Fx)+d(y, Gy)), \ \ \ \hbox{for all} \ x, \ y\in D. \end{equation} (2.27)

    Taking x = 1 and y = 0 in (2.27), we obtain 1\leq \theta , which contradicts \theta\in [0, 1) . Therefore, F and G do not satisfy conditions of Theorem 1.9.

    For any x, y\in D , we have

    \begin{equation*} \begin{aligned} &|x-y|+(|x-Fx|+|y-Gy|)-|Fx-Gy|\\ & = \left\{ \begin{aligned} &|x-y|+|y|-|x|, \ \ \ \hbox{if} \ y\geq 0, \\ &|x-y|-|x-y|, \ \ \ \ \hbox{if} \ y < 0, \end{aligned} \right.\\ &\geq 0, \end{aligned} \end{equation*}

    where the last inequality holds by the triangle inequality of |\cdot| . Therefore, F and G satisfy condition (i) of Theorem 2.5 with \tau = 1 . Let \bar{F}x: = \frac{1}{2}Fx+\frac{1}{2}0 = \frac{1}{2}Fx , \bar{G}x = \frac{1}{2}Gx . Then we can calculate that for all x\in D ,

    \lim\limits_{n\rightarrow \infty}\left\|\bar{F}^{n}x-\bar{F}^{n+1}x\right\| = \lim\limits_{n\rightarrow \infty}\left|\frac{1}{2^{n}}x-\frac{1}{2^{n+1}}x\right| = 0,
    \lim\limits_{n\rightarrow \infty}\left\|\bar{G}^{n}x-\bar{G}^{n+1}x\right\| = \lim\limits_{n\rightarrow \infty}\left\{ \begin{aligned} &0, \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \hbox{if} \ x\geq 0, \\ &\left|\frac{1}{2^{n}}x-\frac{1}{2^{n+1}}x\right|, \ \ \hbox{if} \ x < 0, \end{aligned} \right. = 0.
    \lim\limits_{n\rightarrow \infty}\left\|\bar{F}^{n}x-F(\bar{F}^{n}x)\right\| = \lim\limits_{n\rightarrow \infty}\left|\frac{1}{2^{n}}x-\frac{1}{2^{n}}x\right| = 0,
    \lim\limits_{n\rightarrow \infty}\left\|\bar{G}^{n}x-G(\bar{G}^{n}x)\right\| = \lim\limits_{n\rightarrow \infty}\left\{ \begin{aligned} &0, \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \hbox{if} \ x\geq 0, \\ &\left|\frac{1}{2^{n}}x-\frac{1}{2^{n}}x\right|, \ \ \hbox{if} \ x < 0, \end{aligned} \right. = 0.

    Therefore, conditions (ⅱ) and (ⅲ) of Theorem 2.5 are satisfied. Obviously, F and G are continuous. All conditions of Theorem 2.5 are satisfied. Hence, F and G have a common fixed point. In fact, all points in [-1, 0] are common fixed points of F and G .

    Example 2.8. Let X = \mathbb{R} be equipped with the usual norm |\cdot| , D = \left[0, \frac{1}{2}\right] , and F, \ G: D\rightarrow D be defined by

    \begin{equation*} Fx = x, \ \ \ G(x) = x^{2}, \ \ \ \hbox{for all} \ x\in D. \end{equation*}

    Obviously, the metric induced by the norm is

    d(x, y) = |x-y|, \ \ \ \forall \ x, y\in X.

    We first prove that F and G do not satisfy inequality (1.2) on D . Suppose to the contrary that F and G satisfy

    \begin{equation} d(Fx, Gy)\leq \theta d(x, y)+\tau(d(x, Fx)+d(y, Gy)), \ \ \ \hbox{for all} \ x, \ y\in D. \end{equation} (2.28)

    Taking x = 1 and y = 0 in (2.28), we obtain 1\leq \theta , which contradicts \theta\in [0, 1) . Therefore, F and G do not satisfy the conditions of Theorem 1.9.

    For any x, y\in D , we have

    \begin{equation*} |x-y|+(|x-Fx|+|y-Gy|)-|Fx-Gy| = |x-y|+\left|y-y^{2}\right|-\left|x-y^{2}\right|\geq 0. \end{equation*}

    Thus F and G satisfy condition (ⅰ) of Therem 2.5 with \tau = 1 . Let

    \bar{F}x: = \frac{3}{4}Fx+\frac{1}{4}0 = \frac{3}{4}x, \ \ \ \bar{G}x: = \frac{3}{4}Gx+\frac{1}{4}0 = \frac{3}{4}x^{2}, \ \ \ \hbox{for all } \ x\in D.

    Then we can calculate that for all x\in D ,

    \lim\limits_{n\rightarrow \infty}\left\|\bar{F}^{n}x-\bar{F}^{n+1}x\right\| = \lim\limits_{n\rightarrow \infty}\left|\frac{3^{n}}{4^{n}}x-\frac{3^{n+1}}{4^{n+1}}x\right| = 0,
    \lim\limits_{n\rightarrow \infty}\left\|\bar{G}^{n}x-\bar{G}^{n+1}x\right\| = \lim\limits_{n\rightarrow \infty}\left|\frac{3^{2^{n}-1}}{4^{2^{n}-1}}x^{2^{n}}-\frac{3^{2^{n+1}-1}}{4^{2^{n+1}-1}}x^{2^{n+1}}\right| = 0.
    \lim\limits_{n\rightarrow \infty}\left\|\bar{F}^{n}x-F(\bar{F}^{n}x)\right\| = \lim\limits_{n\rightarrow \infty}\left|\frac{3^{n}}{4^{n}}x-\frac{3^{n}}{4^{n}}x\right| = 0,
    \lim\limits_{n\rightarrow \infty}\left\|\bar{G}^{n}x-G(\bar{G}^{n}x)\right\| = \lim\limits_{n\rightarrow \infty}\left|\frac{3^{2^{n}-1}}{4^{2^{n}-1}}x^{2^{n}}-\left(\frac{3^{2^{n}-1}}{4^{2^{n}-1}}x^{2^{n}}\right)^{2}\right| = 0.

    Therefore, F and G satisfy conditions (ⅱ) and (ⅲ) of Theorem 2.5. Clearly, F and G satisfy condition (ⅳ) since they are continuous on D . Now, all conditions of Theorem 2.5 are satisfied, and so F and G have a common fixed point. In fact, u = 0 is the unique common fixed point of F and G on X .

    In this paper, following the argument of [14], we studied the common fixed point of two self-mappings. In this direction, most of results in the literature required that the contractive coefficient is constant and in interval [0, 1) . However, in some mathematical models, a fixed contractive coefficient in [0, 1) does not exist. In order to overcome this difficulty, we replaced the contractive coefficient with a contractive function with its range in [0, 1] . By using the concepts of asymptotical regularity, orbital continuity and q -continuity, and some techniques of mathematical analysis, a unique common fixed point theorem for two self-mappings has been established in a metric space. Furthermore, for a pair of nonexpansive mappings, by constructing two approximate mappings of them, the concepts of asymptotical regularity and continuity, a common fixed point has been derived. We also presented some examples to illustrate that our results extended the main result in [14]. The obtained results will bring potential applications in the existence of solution of some mathematical models.

    In future studies, we will extend our results to more general enriched Banach contractions mappings in a Banach space, and obtain an iterative sequence, which converges to a common fixed point of them. Moreover, following the idea of our previous work [13], we will consider the common fixed point of finite self-mappings without contraction conditions in a Banach space, and its applications in variational inequalities.

    The author sincerely thanks the anonymous reviewers for their insightful comments which greatly improved the quality of the paper. This work was supported by the National Natural Science Foundation of China (12061085) and the Applied Basic Research Project of Yunnan Province (202001BB050036).

    The authors declare that they have no competing interests.



    [1] P. Agarwal, M. Jleli, B. Samet, Fixed point theory in metric spaces, Springer, 2018. https://doi.org/10.1007/978-981-13-2913-5
    [2] R. K. Bisht, A note on the fixed point theorem of G\acute{\hbox{o}}rnicki, J. Fix. Point Theory A., 21 (2019), 54. https://doi.org/10.1007/s11784-019-0695-x doi: 10.1007/s11784-019-0695-x
    [3] R. K. Bisht, N. K. Singh, On asymptotic regularity and common fixed points, J. Anal., 28 (2020), 847–852. https://doi.org/10.1007/s41478-019-00213-0 doi: 10.1007/s41478-019-00213-0
    [4] N. Boonsatit, G. Rajchakit, R. Sriraman, C. P. Lim, P. Agarwal, Finite-/fixed-time synchronization of delayed Clifford-valued recurrent neural networks, Adv. Differ. Equ., 2021 (2021), 276. https://doi.org/10.1186/s13662-021-03438-1 doi: 10.1186/s13662-021-03438-1
    [5] F. E. Browder, W. V. Petryshyn, The solution by iteration of nonlinear functional equations in Banach spaces, Bull. Am. Math. Soc., 72 (1966), 571–575. https://doi.org/10.1090/S0002-9904-1966-11544-6 doi: 10.1090/S0002-9904-1966-11544-6
    [6] S. Carl, S. Heikkil\ddot{\hbox{a}}, Fixed point theory in ordered sets and applications, Springer, 2011. https://doi.org/10.1007/978-1-4419-7585-0
    [7] L. Ćirić, On contraction type mappings, Math. Balk., 1 (1971), 52–57.
    [8] M. A. Geraghty, On contractive mappings, Proc. Am. Math. Soc., 40 (1973), 604–608. https://doi.org/10.1090/S0002-9939-1973-0334176-5 doi: 10.1090/S0002-9939-1973-0334176-5
    [9] J. Górnicki, Remarks on asymptotic regularity and fixed points, J. Fix. Point Theory A., 21 (2019), 29. https://doi.org/10.1007/s11784-019-0668-0 doi: 10.1007/s11784-019-0668-0
    [10] A. Granas, J. Dugundji, Fixed point theory, Springer, 2003. https://doi.org/10.1007/978-0-387-21593-8
    [11] H. A. Hammad, P. Agarwal, J. L. G. Guirao, Applications to boundary value problems and homotopy theory via tripled fixed point techniques in partially metric spaces, Math., 9 (2021), 1–22. https://doi.org/10.3390/math9162012 doi: 10.3390/math9162012
    [12] S. Hassan, M. D. Sen, P. Agarwal, Q. Ali, A. Hussain, A new faster iterative scheme for numerical fixed points estimation of Suzuki's generalized nonexpansive mappings, Math. Probl. Eng., 2020 (2020), 1–9. https://doi.org/10.1155/2020/3863819 doi: 10.1155/2020/3863819
    [13] H. Huang, J. Zou, A note on the paper "a common fixed point theorem with applications", J. Optimiz. Theory Appl., 168 (2016), 1087–1090. https://doi.org/10.1007/s10957-015-0836-3 doi: 10.1007/s10957-015-0836-3
    [14] A. R. Khan, D. M. Oyetunbi, On some mappings with a unique common fixed point, J. Fix. Point Theory A., 22 (2020), 47. https://doi.org/10.1007/s11784-020-00781-w doi: 10.1007/s11784-020-00781-w
    [15] A. Pant, R. P. Pant, Fixed points and continuity of contractive maps, Filomat, 31 (2017), 3501–3506. https://doi.org/10.2298/FIL1711501P doi: 10.2298/FIL1711501P
    [16] T. Rasham, P. Agarwal, L. S. Abbasi, S. Jain, A study of some new multivalued fixed point results in a modular like metric space with graph, J. Anal., 30 (2022), 833–844. https://doi.org/10.1007/s41478-021-00372-z doi: 10.1007/s41478-021-00372-z
    [17] B. Wang, H. Jahanshahi, C. Volos, S. Bekiros, A. Yusuf, P. Agarwal, A. A. Aly, Control of a symmetric chaotic supply chain system using a new fixed-time super-twisting sliding mode technique subject to control input limitations, Symmetry, 13 (2021), 1257. https://doi.org/10.3390/sym13071257 doi: 10.3390/sym13071257
    [18] H. K. Xu, An iterative approach to quadratic optimization, J. Optimiz. Theory Appl., 116 (2003), 659–678. https://doi.org/10.1023/A:1023073621589 doi: 10.1023/A:1023073621589
  • This article has been cited by:

    1. Godwin Chidi Ugwunnadi, Abdul Rahim Khan, Austine Efut Ofem, Convergence results for Górnicki type contractive mapping in CAT(0) spaces, 2024, 0971-3611, 10.1007/s41478-024-00864-8
  • Reader Comments
  • © 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(1880) PDF downloads(149) Cited by(1)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog