Obstacle Avoidance and Tracking Control of Redundant Robotic Manipulator: An RNN-Based Metaheuristic Approach

Khan, Ameer Hamza; Li, Shuai; Luo, Xin

doi:10.1109/tii.2019.2941916

Cited by 180 publications

(76 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The estimation of an approximated gradient is a key feature of BAS, which distinguishes it from another metaheuristic algorithm. Since its introduction, BAS has found its application in several real-world systems [17], [55]- [64]. The working of the original BAS can be described like this; at each iteration, the value of the objective function is computed at each antennae fiber location, a vector is drawn from the fiber with the lowest value toward the fiber with the highest value, the vector represents the direction of the approximated gradient.…”

Section: (A)mentioning

confidence: 99%

See 1 more Smart Citation

BAS-ADAM: an ADAM based approach to improve the performance of beetle antennae search optimizer

Khan

Cao

et al. 2020

IEEE/CAA J. Autom. Sinica

Self Cite

152

View full text Add to dashboard Cite

In this paper, we propose enhancements to Beetle Antennae Search (BAS) algorithm, called BAS-ADAM, to smoothen the convergence behavior and avoid trapping in local-minima for a highly non-convex objective function. We achieve this by adaptively adjusting the step-size in each iteration using the Adaptive Moment Estimation (ADAM) update rule. The proposed algorithm also increases the convergence rate in a narrow valley. A key feature of the ADAM update rule is the ability to adjust the step-size for each dimension separately instead of using the same step-size. Since ADAM is traditionally used with gradient-based optimization algorithms, therefore we first propose a gradient estimation model without the need to differentiate the objective function. Resultantly, it demonstrates excellent performance and fast convergence rate in searching for the optimum of non-convex functions. The efficiency of the proposed algorithm was tested on three different benchmark problems, including the training of a high-dimensional neural network. The performance is compared with Particle Swarm Optimizer (PSO) and the original BAS algorithm.

show abstract

Section: (A)mentioning

confidence: 99%

“…Such conditions do not hold for the vast majority of the systems, e.g., integer programming [14]. In fact, for the vast majority of the optimization problem, specifically in the control system [15], an accurate model of the system might be unknown in advance and require real-time estimation of parameters [16], [17].…”

Section: Introductionmentioning

confidence: 99%

BAS-ADAM: an ADAM based approach to improve the performance of beetle antennae search optimizer

Khan

Cao

et al. 2020

IEEE/CAA J. Autom. Sinica

Self Cite

152

View full text Add to dashboard Cite

show abstract

“…For the step size parameter δ(t) and the detection distance parameter λ(t), they often need to be appropriately chosen through necessary optimal computation. After absorbing some of the experience of BAS-WPT [38] and BAORNN [32], we define the updating law for δ(t) and λ(t) as follows…”

Section: Step-size and Antenna Exploration Range Parameter Updatingmentioning

confidence: 99%

“…BAORNN (beetle antennae olfactory recurrent neural network) algorithm focuses on the combination of BAS and recurrent neural network, so as to achieve motion planning of redundant robotic arms in three dimensions. BAORNN also solves problems from the velocity layer [32]. The basic principle of the BAS algorithm is as follows.…”

Section: Introductionmentioning

confidence: 99%

Motion Planning of Redundant Manipulator With Variable Joint Velocity Limit Based on Beetle Antennae Search Algorithm

Cheng

et al. 2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Redundant manipulators play important roles in many industrial and service applications by assisting people fulfill heavy and repetitive jobs. However, redundant manipulators are coupled highlynonlinear systems which exert difficulty of redundancy resolution computation. Conventional methods such as pseudo-inverse-based approaches obtain the resolved joint angles from joint velocity level, which may bring about more computational cost and may neglect joint velocity limits. In this work, a motion planning method based on beetle antennae search algorithm (BAS) is proposed for motion planning of redundant manipulators with the variable joint velocity limit. Such proposed work does not need to resolve the velocity kinematics equation as the conventional methods do, and the proposed method can directly deal with the forward kinematics equation to resolve the desired joint angles. The simulation and experiment on the five-link planar manipulator and the Kuka industrial manipulator system demonstrate the efficiency of the proposed method for motion planning of redundant manipulator, and reveal the reliable performance of the BAS algorithm as compared with genetic algorithm (GA), particle swarm optimization (PSO), firefly algorithm(FA) and quantum behaved particle swarm algorithm(QPSO) methods.

show abstract

“…In this searching mechanism, the algorithm only needs to know the objective function and the searching scope, and the aim solution can be obtained whether the searching scope is continuously derivable or not. Therefore, Meta-heuristic algorithms are widely applied in many fields, such as obstacle avoidance [11], automatic control [12], scheduling problems [13], image evolution [14], path planning [15], information feedback models [16] and so on. Meta-heuristic algorithms are mainly divided into three parts, including the biological evolutionary method, the species living habit, and the nature phenomena.…”

mentioning

confidence: 99%

A Self-Adaption Butterfly Optimization Algorithm for Numerical Optimization Problems

et al. 2020

View full text Add to dashboard Cite

For shortcomings of poor exploaration and parameter complexities of the butterfly optimization algorithm, an improved butterfly optimization algorithm based the self-adaption method (SABOA) was proposed to extremely enhance the searching accuracy and the iteration capability. SABOA has advantages of having fewer parameters, the simple algorithm structure, and the strong precision. First, a new fragrance coefficient was added to the basic butterfly optimization algorithm. Then, new iteration and updating strategies were introduced in global searching and local searching phases. Finally, this paper tested different optimization problems including low-high functions and constrained problems, and the obtained results were compared with other well-known algorithms, this paper also drew various mathematical statistics figures to comprehensively analyze searching performances of the proposed algorithms. The experimental results show that SABOA can get less number of function evaluations compared to other considered algorithms, which illustrates that SABOA has great searching balance, large exploration, and high iterative speed. INDEX TERMS Butterfly optimization algorithm, global optimization, constrained problem. I. INTRODUCTION In applied mathematics and engineering fields, there are numerous optimization problems whose calculated solutions are in a large and complex searching space. Traditional optimization methods, including the steepest descent method, the conjugate gradient method, the variable scale method, and Newton method, can only deal with objective functions that are simple, continuously differentiable, and high order differentiable [1]-[3]. With the increasing of problem diversities and problem complexities, traditional optimization methods can not meet different requirements of higher calculation speed and lower average percentage error, so it is crucial to find for new optimization methods that have fast calculation speed and perfect convergence abilities [4], [5]. With the development of artificial intelligence, digitization, and computer technologies, numerous meta-heuristic optimization algorithms have been increasingly proposed and applied in science and engineering fields [6]-[8]. Meta-heuristic algorithms own characteristics of selforganizing, mutual compatibility, simplicity, parallelism, wholeness and harmony. Meta-heuristic algorithms work The associate editor coordinating the review of this manuscript and approving it for publication was Jagdish Chand Bansal.

show abstract

Obstacle Avoidance and Tracking Control of Redundant Robotic Manipulator: An RNN-Based Metaheuristic Approach

Cited by 180 publications

References 41 publications

BAS-ADAM: an ADAM based approach to improve the performance of beetle antennae search optimizer

BAS-ADAM: an ADAM based approach to improve the performance of beetle antennae search optimizer

Motion Planning of Redundant Manipulator With Variable Joint Velocity Limit Based on Beetle Antennae Search Algorithm

A Self-Adaption Butterfly Optimization Algorithm for Numerical Optimization Problems

Contact Info

Product

Resources

About