Tangent navigated robot path planning strategy using particle swarm optimized artificial potential field

doi:10.1016/j.ijleo.2017.12.169

Optik

Volume 158, April 2018, Pages 639-651

https://doi.org/10.1016/j.ijleo.2017.12.169 Get rights and content

Abstract

Artificial potential field has long been proposed in the field of robot path planning. But the well-known drawbacks like local minimal problem and low efficiency prevent its wide application. In this paper, we propose a particle swarm optimized tangent vector based artificial potential field path planning algorithm (PSO-TVAPF) to solve those problems. A tangent vector based on obstacles’ information is added into artificial potential field (APF) model as an auxiliary force for obstacle avoiding process. This makes the original model, tangent vector based artificial potential field (TVAPF). To achieve more intelligent and efficient TVAPF, map and path information are taking into consideration dynamically while calculating tangent vector. In addition, particle swarm optimization has been used to refine TVAPF, which leads to the final model named PSO-TVAPF. Simulation experiments and physical validation results indicate that the proposed algorithm can overcome classic APF's drawbacks and improve path planning efficiency significantly.

Introduction

Robot path planning [1], [2], [3] is an important research field in robot automation and is also the foundation of quantities of robot tasks. It has been proved that they can help human beings in many works, such as handling cargos, cleaning, inspecting, autonomous underwater vehicle research [4], [5], etc. The wide application and bright prospect makes path planning of wheel mobile robot an important research field.

In this paper, we propose a new approach called tangent vector based artificial potential field (TVAPF) and it is optimized model using particle swarm optimization (PSO-TVAPF) for path planning in mobile robots. The TVAPF proposal is based on artificial potential field (APF) enhanced with a tangent vector which figures out according to obstacles interactively. The path planning with TVAPF proposal consider the obstacle avoiding problem at its beginning different from most APF algorithm they start to go round obstacle at a very close distance. By this way, our algorithm eliminate local minimal problem and shorten total path length. To achieve more intelligent and efficient TVAPF, map and path information are taking into consideration dynamically while calculating tangent vector. In addition, particle swarm optimization has been used to refine parameters of TVAPF.

The rest parts of this paper are organized as follow. In Section 2, some basic theories and previous research results related to our works would be introduced. In Section 3, we give a detailed description about our proposed algorithms, TVAPF and PSO-TVAPF, in robot path planning. Experiment results presents in Section 4, where we first compares four RPP approaches, APF, BPF, TVAPF and PSO-TVAPF on a MFC simulation program and then validate our algorithm on a wheel mobile robot produced by Ingenious Corporation. At last, a conclusion is given to summarize our algorithm in this paper.

Section snippets

Related works

Robot motion planning launched at mid-1960s, but it was not until Lozano-Prezs revolutionary contribution on spatial planning that MP drew most researchers’ attention. In the past few decades, researchers in the field of mobile robot path planning have put forwarded many algorithms, which has been dominated by classical approaches such as the roadmap, cell decomposition and artificial potential field (APF). Representative proposals of roadmaps approaches are the visibility graph which is a

Tangent vector based artificial potential field and its particle swarm optimized model

Definition 1

The line from current position P_c to current goal P_g is navigation line L_cg.

Definition 2

D_p is the distance from an outside point P to navigation line L_cg.

(9) $D_{p} = \frac{2 S_{Δ}}{D_{CG}}$

Where S_Δ is the area of triangle consist of point P, P_c and P_g, D_cg is distance from P_c to P_g.

D_{p} = \frac{2 S_{Δ}}{D_{CG}}

Definition 3

η is path planning efficiency,

(10) $η = \frac{D (Start, End)}{D_{Actual} (Start, End)}$

D_Actual(Start, End) is actual distance given by path planning algorithms, D(Start, End) is direct distance between start and goal.

η = \frac{D (Start, End)}{D_{Actual} (Start, End)}

Simulations and experiments results

In this section, three set of experiments would be given to validate the proposed algorithms in Section 3 and analyze the promotion compare to traditional APF. We mainly focus on the ability of avoiding local minimal problem and the shortening rate on path length. The first two group of experiments are completed in an simulation environment construct of MFC program by visual studio 2015 on a personal computer, configured as follow: Intel Core i7-4720HQ CPU @2.60 GHz; 8 GB RAM; 64 bit Windows 10

Conclusion

In this paper, we proposed a new path planning algorithm, tangent vector based artificial potential field (TVAPF), based on artificial potential field. TVAPF calculates tangent vector of obstacles before obstacle avoiding process and combine it with potential field force to drive robot approach the goal. In addition, we use PSO to optimize our proposed TVAPF model. Experiments results indicates that our proposed TVAPF and PSO-TVAPF algorithm can effectively eliminate local minimal problems and

Acknowledgements

This work is supported by Zhejiang Provincial Natural Science Foundation of China (No. LY18F030018) and Natural Science Foundation of China (No. 51376055).

References (21)

H. Cao et al.
Visualized trajectory planning of flexible redundant robotic arm using a novel hybrid algorithm
Optik
(2016)
B. Zhang et al.
A path planning strategy for searching the most reliable path in uncertain environments
Int. J. Adv. Robot. Syst.
(2016)
F. Kamil et al.
New robot navigation algorithm for arbitrary unknown dynamic environments based on future prediction and priority behavior
Expert Syst. Appl.
(2017)
W. Nie et al.
Planar path-following tracking control for an autonomous underwater vehicle in the horizontal plane
Optik
(2016)
O. Montiel et al.
Path planning for mobile robots using bacterial potential field for avoiding static and dynamic obstacles
Expert Syst. Appl.
(2015)
B. He et al.
Autonomous navigation based on unscented-FastSLAM using particle swarm optimization for autonomous underwater vehicles
Measurement
(2015)
M.J. Mahmoodabadi et al.
Optimal robust sliding mode tracking control of a biped robot based on ingenious multi-objective PSO
Neurocomputing
(2014)
A.T. Rashid et al.
Path planning with obstacle avoidance based on visibility binary tree algorithm
Robot. Auton. Syst.
(2013)
B. He et al.
A distributed parallel motion control for the multi-thruster autonomous underwater vehicle
Mech. Des. Struct. Mach.
(2013)
J. Berger et al.
A hybrid genetic algorithm for rescue path planning in uncertain adversarial environment
IEEE Congress Evol. Comput.
(2010)

There are more references available in the full text version of this article.

Cited by (79)

Finite-time adaptive event-triggered control for USV with COLREGS-compliant collision avoidance mechanism
2023, Ocean Engineering
This paper focuses on the real time intelligent navigation scheme for unmanned surface vehicle (USV) subject to collision avoidance regulations. The proposed approach is formulated as two segments: the real time guidance principle with COLREGS-compliant collision avoidance mechanism and the finite-time adaptive event-triggered control algorithm. The pointed improvement for the artificial potential field (APF) can guarantee USV possess the excellent avoidance function in accordance with COLREGS regulations. Especially, when the obstacle makes dangerous actions, the emergency obstacle avoidance operation (EOAO) is implemented according to the real time position of obstacles. Based on the proposed guidance principle, a finite-time robust neural control algorithm is developed by fusion of the finite-time stability theory and event-triggered mechanism. The intermittent computing and communication can effectively reduce the energy and resource consumption of the control system. Moreover, the finite-time adaptive controller design contributes to smaller computation load and facilitates the implementation of the algorithm in ocean engineering. The semi global finite-time uniformly boundedness (SGFTUB) of the closed-loop system is guaranteed through the Lyapunov theory. Finally, the numerical simulation and comparison experiment are illustrated to verify the guidance and control performance of the proposed algorithm.
Near-optimal 3D trajectory design in presence of obstacles: A convolutional neural network approach
2023, Robotics and Autonomous Systems
This paper proposes an approach based on neural networks for designing near-optimal 3D trajectories connecting two points separated by obstacles. A reference path is first built with a novel Theta* algorithm implementation, upgraded to reduce the number of waypoints and the angular variations. Starting from the path, a trajectory based on piecewise Bézier curves is designed with an algorithm relying on two design parameters. The optimal value of these parameters is estimated with a Convolutional Neural Network (CNN) architecture receiving as inputs an image of the path and its properties. The CNN training is performed on a synthetic dataset of optimal parameters built with Differential Evolution optimization for a variety of randomly generated paths. The parameters from CNN are refined with an algorithm capable of providing with high success rate near-optimal trajectories within minimum computation time.
A non-potential orthogonal vector field method for more efficient robot navigation and control
2023, Robotics and Autonomous Systems
Citation Excerpt :
However, these methods rarely consider the theoretical completeness after the modification. The method in [23] actually changes the gradient vector field of the APF method into non-potential without stability proof. Besides, in some literature, the APF method is also used as a baseline in the efficiency comparison.
To navigate and control a single mobile robot or a robotic swarm with higher efficiency, a novel non-potential orthogonal vector field method is proposed in this paper, which is modified from the traditional artificial potential field method. The improvement strategy aims at making the overall repulsive vector field orthogonal to the attractive vector field in some conditions. And the same potential field function is still applied to the Lyapunov-based stability analysis. In short, such an improvement strategy combines the advantages of the artificial potential field method with the non-potential vector field method, namely a combination of theoretical completeness and control efficiency. Finally, the effectiveness of the proposed method is validated both by numerical simulations with statistical significance and real experiments. The comparisons between our method and other methods are also presented.
A hybrid formation path planning based on A* and multi-target improved artificial potential field algorithm in the 2D random environments
2022, Advanced Engineering Informatics
Citation Excerpt :
However, the algorithm is easy to make the robot fall into local minimum, which can cause that the target point is unreachable or the robot occur oscillation. In order to solve the problem, the literature [20] proposes the tangent vector based artificial potential field (TVAPF) algorithm by using obstacles tangent points as the virtual sub-target points of APF algorithm. But this algorithm requires the robot to move to the position of the tangent points every time, so there are problems of long redundant paths and less optimal planning path.
Traditional artificial potential field algorithm for multi-robot formation is easy to fall into local minima and the path planning efficiency is low. To this end, we propose a new method of a hybrid formation path planning based on A* and multi-target improved artificial potential field algorithm (A*-MTIAPF) that provides the optimal collision free path and improves the efficiency for multi-robot formation path planning. The A*-MTIAPF algorithm integrates global path planning and local path planning. The novelties include combining A* with the improved artificial potential field algorithm and dividing multiple virtual sub-target points on the global optimal path of A* planning. Firstly, A* algorithm is used to complete the global path planning. Secondly, the improved artificial potential field algorithm which takes multiple sub-target points divided by the global optimal path as virtual target points is used to complete local path planning by switching target points. In addition, we propose a double priority judgment control algorithm (DPJC) to solve the collision problem among multiple robots by setting double priority to determine the movement order of each robot. Then, a new experimental method is designed by using the randomly generated 2D maps to verify the effectiveness of the proposed method. The results show that our method has advantages that it solves the local minimum problem, improves the efficiency of formation path planning and avoids collision among multiple robots over existing methods.
Combined grid and heat conduction optimization for staircase cleaning robot path planning
2022, Automation in Construction
Citation Excerpt :
However, the obtained paths tend to be local optima, and applications to higher dimensional problems are limited. In the group of virtual potential field (VPF) methods [28–31], the robot configuration space free domain is represented as a potential field where the target position and obstacles are modeled as attractive and repulsive potentials, respectively. The advantages of VPF methods lie in the fact that they can react rapidly to dynamic environments and are able to deal with higher-dimensional PP problems.
Staircase climbing, particularly in complex environments of multi-storey buildings, is a challenging task for robotics. By leveraging the similarities between robot path planning and heat conduction, this paper presents a novel path planning to direct with optimal energy consumption the self-reconfigurable staircase cleaning robot called sTetro, able to transform its shape to perform both descent and ascent motions on the staircase. Finding optimal path is modeled as heat traveling between heat source and heat sink through thermal conductive materials. We utilize the temperature gradient on grid-based optimization method to search for optimal paths by minimizing path length and energy consumption. The proposed path planning method successfully applies to six virtual environments with various obstacle placements. Additionally, validation tests on a real sTetro robot in 2 real staircase scenarios show that the energy consumption of the optimized paths from the proposed method is about 21% lower than that from conventional methods.
Multi-Objective Evolutionary Artificial Potential Field for Indoor Path Planning
2024, International Journal of Intelligent Systems and Applications in Engineering

View all citing articles on Scopus

View full text

Original research articleTangent navigated robot path planning strategy using particle swarm optimized artificial potential field

Abstract

Introduction

Section snippets

Related works

Tangent vector based artificial potential field and its particle swarm optimized model

Simulations and experiments results

Conclusion

Acknowledgements

Optik

Int. J. Adv. Robot. Syst.

Expert Syst. Appl.

Optik

Expert Syst. Appl.

Measurement

Neurocomputing

Robot. Auton. Syst.

A distributed parallel motion control for the multi-thruster autonomous underwater vehicle

Mech. Des. Struct. Mach.

A hybrid genetic algorithm for rescue path planning in uncertain adversarial environment

IEEE Congress Evol. Comput.

Original research article
Tangent navigated robot path planning strategy using particle swarm optimized artificial potential field