scholarly journals Deep Reinforcement Learning-Based Path Planning for Multi-Arm Manipulators with Periodically Moving Obstacles

2021 ◽  
Vol 11 (6) ◽  
pp. 2587
Author(s):  
Evan Prianto ◽  
Jae-Han Park ◽  
Ji-Hun Bae ◽  
Jung-Su Kim
Keyword(s):  
Path Planning ◽  
Optimal Path ◽  
Training Data ◽  
Planning Problem ◽  
Position Information ◽  
Moving Obstacles ◽  
Dimensional Property ◽  
Experience Replay ◽  
Planning Algorithm ◽  
Finite Time Horizon

In the workspace of robot manipulators in practice, it is common that there are both static and periodic moving obstacles. Existing results in the literature have been focusing mainly on the static obstacles. This paper is concerned with multi-arm manipulators with periodically moving obstacles. Due to the high-dimensional property and the moving obstacles, existing results suffer from finding the optimal path for given arbitrary starting and goal points. To solve the path planning problem, this paper presents a SAC-based (Soft actor–critic) path planning algorithm for multi-arm manipulators with periodically moving obstacles. In particular, the deep neural networks in the SAC are designed such that they utilize the position information of the moving obstacles over the past finite time horizon. In addition, the hindsight experience replay (HER) technique is employed to use the training data efficiently. In order to show the performance of the proposed SAC-based path planning, both simulation and experiment results using open manipulators are given.

scholarly journals Review on path planning algorithm for unmanned aerial vehicles

2021 ◽  
Vol 24 (2) ◽  
pp. 1017
Author(s):  
Nurul Saliha Amani Ibrahim ◽  
Faiz Asraf Saparudin
Keyword(s):  
Path Planning ◽  
Unmanned Aerial Vehicles ◽  
Autonomous Vehicles ◽  
Autonomous Vehicle ◽  
Optimal Path ◽  
Computation Time ◽  
Planning Problem ◽  
Aerial Vehicles ◽  
Planning Algorithm ◽  
Path Planning Algorithm

The path planning problem has been a crucial topic to be solved in autonomous vehicles. Path planning consists operations to find the route that passes through all of the points of interest in a given area. Several algorithms have been proposed and outlined in the various literature for the path planning of autonomous vehicle especially for unmanned aerial vehicles (UAV). The algorithms are not guaranteed to give full performance in each path planning cases but each one of them has their own specification which makes them suitable in sophisticated situation. This review paper evaluates several possible different path planning approaches of UAVs in terms optimal path, probabilistic completeness and computation time along with their application in specific problems.

Towards Design of an Efficient Sensing Data Acquisition scheme for UAVs-assisted Wireless Sensor Networks

2022 ◽  
Vol 13 (2) ◽  
pp. 0-0
Keyword(s):  
Path Planning ◽  
Data Acquisition ◽  
Large Scale ◽  
Cluster Formation ◽  
Optimal Path ◽  
Research Work ◽  
Planning Problem ◽  
Sensing Data ◽  
Planning Algorithm ◽  
Load Balanced

This paper investigates sensing data acquisition issues from large-scale hazardous environments using UAVs-assisted WSNs. Most of the existing schemes suffer from low scalability, high latency, low throughput, and low service time of the deployed network. To overcome these issues, we considered a clustered WSN architecture in which multiple UAVs are dispatched with assigned path knowledge for sensing data acquisition from each cluster heads (CHs) of the network. This paper first presents a non-cooperative Game Theory (GT)-based CHs selection algorithm and load balanced cluster formation scheme. Next, to provide timely delivery of sensing information using UAVs, hybrid meta-heuristic based optimal path planning algorithm is proposed by combing the best features of Dolphin Echolocation and Crow Search meta-heuristic techniques. In this research work, a novel objective function is formulated for both load-balanced CHs selection and for optimal the path planning problem. Results analyses demonstrate that the proposed scheme significantly performs better than the state-of-art schemes.

scholarly journals Local path planning for mobile robots based on intermediate objectives

Robotica ◽  
2014 ◽  
Vol 33 (4) ◽  
pp. 1017-1031 ◽  
Author(s):  
Yingchong Ma ◽  
Gang Zheng ◽  
Wilfrid Perruquetti ◽  
Zhaopeng Qiu
Keyword(s):  
Path Planning ◽  
Mobile Robots ◽  
Autonomous Navigation ◽  
Optimal Path ◽  
Kinematic Model ◽  
Planning Problem ◽  
Planning Algorithm ◽  
Local Path ◽  
Velocity Constraints ◽  
Horizon Planning

SUMMARYThis paper presents a path planning algorithm for autonomous navigation of non-holonomic mobile robots in complex environments. The irregular contour of obstacles is represented by segments. The goal of the robot is to move towards a known target while avoiding obstacles. The velocity constraints, robot kinematic model and non-holonomic constraint are considered in the problem. The optimal path planning problem is formulated as a constrained receding horizon planning problem and the trajectory is obtained by solving an optimal control problem with constraints. Local minima are avoided by choosing intermediate objectives based on the real-time environment.

scholarly journals Path Planning Algorithm for Mobile Robot Based on Improved Ant Colony Algorithm

2021 ◽  
Vol 2083 (4) ◽  
pp. 042033
Author(s):  
Yanyun Li ◽  
Fenggang Liu
Keyword(s):  
Path Planning ◽  
Mobile Robot ◽  
Path Length ◽  
Ant Colony Algorithm ◽  
Planning Process ◽  
Optimal Path ◽  
Ant Colony ◽  
Position Information ◽  
Environment Model ◽  
Planning Algorithm

Abstract Due to the influence of full traversal environment, the path length obtained by existing methods is too long. In order to improve the performance of path planning and obtain the optimal path, a full traversal path planning method for omnidirectional mobile robots based on ant colony algorithm is proposed. On the basis of the topology modeling schematic diagram, according to the position information of the mobile robot in the original coordinate system, a new environment model is established by using the Angle transformation. Considering the existing problems of ant colony algorithm, the decline coefficient is introduced into the heuristic function to update the local pheromone, and the volatility coefficient of the pheromone is adjusted by setting the iteration threshold. Finally, through the design of path planning process, the planning of omnidirectional mobile robot’s full traversal path is realized. Experimental results show that the proposed method can not only shorten the full traversal path length, but also shorten the time of path planning to obtain the optimal path, thus improving the performance of full traversal path planning of omnidirectional mobile robot.

scholarly journals Research on Trajectory Planning in Thunderstorm Weather Based on Dynamic Window Algorithm during Approach Segment

2022 ◽  
Vol 2022 ◽  
pp. 1-10
Author(s):  
Li Lu ◽  
Chenyu Liu
Keyword(s):  
Path Planning ◽  
Optimal Path ◽  
Dynamic Environment ◽  
Time Correlation ◽  
Space Time ◽  
Superior Performance ◽  
Time Dynamic ◽  
Moving Obstacles ◽  
Planning Algorithm ◽  
Window Approach

Dynamic window algorithm (DWA) is a local path-planning algorithm, which can be used for obstacle avoidance through speed selection and obtain the optimal path, but the algorithm mainly plans the path for fixed obstacles. Based on DWA algorithm, this paper proposes an improved DWA algorithm based on space-time correlation, namely, space-time dynamic window approach. In SDWA algorithm, a DWA associated with obstacle position and time is proposed to achieve the purpose of path planning for moving obstacles. Then, by setting the coordinates of the initial moving obstacle and identifying safety distance, we can define the shape of the obstacle and the path planning of the approach segment in thunderstorm weather based on the SDWA model was realized. Finally, the superior performance of the model was verified by setting moving obstacles for path planning and selecting the aircraft approach segment in actual thunderstorm weather. The results showed that SDWA has good path-planning performance in a dynamic environment. Its path-planning results were very similar to an actual aircraft performing thunderstorm-avoidance maneuvers, but with more smooth and economical trajectory. The proposed SDWA model had great decision-making potential for approach segment planning in thunderstorm weather.

scholarly journals An Efficient Path Planning Algorithm for Networked Robots using Modified Optimization Algorithm

2019 ◽  
Vol 8 (12) ◽  
pp. 2738-2743
Keyword(s):  
Path Planning ◽  
Optimization Algorithm ◽  
Optimal Path ◽  
Object Identification ◽  
Planning Problem ◽  
Networked Robots ◽  
Design Choice ◽  
Planning Algorithm ◽  
3D Cnn ◽  
Path Planning Algorithm

Path planning has played a significant role in major numerous decision-making techniques through an automatic system involved in numerous military applications. In the last century, pathfinding and generation were carried out by multiple intelligent approaches. It is very difficult in pathfinding to reduce energy. Besides suggesting the shortest path, it has been found that optimal path planning. This paper introduces an efficient path planning algorithm for networked robots using modified optimization algorithms in combination with the η3 -splines. A new method has employed a cuckoo optimization algorithm to handle the mobile robot path planning problem. At first, η3 - splines are combined so an irregular set of points can be included alongside the kinematic parameters chosen to relate with the development and the control of mobile robots. The proposed algorithm comprises of adaptive random fluctuations (ARFs), which help to deal with the very much manageable neighborhood convergence. This algorithm carries out the process of accurate object identification along with analyzing the influence of different design choice by developing a 3D CNN architecture to determine its performance. Besides offering classification in real-time applications, the proposed algorithm outperforms the performance of state of the art in different benchmarks

A Multi-Objective Path Planning Algorithm for Mobile Robots Based on Cellular Automata

2015 ◽  
Author(s):  
Yousef Naranjani ◽  
Jian-Qiao Sun
Keyword(s):  
Path Planning ◽  
Optimization Problems ◽  
Optimal Path ◽  
Planning Problem ◽  
Robot Path Planning ◽  
Maximum Slope ◽  
Multi Objective ◽  
Planning Algorithm ◽  
Elevation Changes ◽  
Robot Path

Many real-world applications of robot path planning involves not only finding the shortest path, but also achieving some other objectives such as minimizing fuel consumption or avoiding danger areas. This paper introduces a 2D path planning scheme that solves a multi-objective path planning problem on a 3D terrain. This allows the controller to pick the most suitable path among a set of optimal paths. The algorithm generates a cellular automaton for the terrain based on the objectives by applying various weighting factors via an evolutionary algorithm and finds the optimal path between the start point and the goal for each set of parameters considering static obstacles and maximum slope constraints. All the final trajectories share the same characteristic that they are non-dominated with respect to the rest of the set in the Multi-Objective Optimization Problems (MOP) context. The objectives considered in this study includes the path length, the elevation changes and avoiding the radars. Testing the algorithm on several problems showed that the method is very promising for mobile robot path planning applications.

Method of evolving junctions: A new approach to optimal path-planning in 2D environments with moving obstacles

2017 ◽  
Vol 36 (4) ◽  
pp. 403-413 ◽  
Author(s):  
Wuchen Li ◽  
Shui-Nee Chow ◽  
Magnus Egerstedt ◽  
Jun Lu ◽  
Haomin Zho
Keyword(s):  
Path Planning ◽  
Optimal Path ◽  
Optimal Solution ◽  
Optimization Method ◽  
Planning Problem ◽  
Optimal Path Planning ◽  
Moving Obstacles ◽  
Finite Dimensional ◽  
General Convex ◽  
Global Optimal

We propose a novel algorithm to find the global optimal path in 2D environments with moving obstacles, where the optimality is understood relative to a general convex continuous running cost. By leveraging the geometric structures of optimal solutions and using gradient flows, we convert the path-planning problem into a system of finite dimensional ordinary differential equations, whose dimensions change dynamically. Then a stochastic differential equation based optimization method, called intermittent diffusion, is employed to obtain the global optimal solution. We demonstrate, via numerical examples, that the new algorithm can solve the problem efficiently.

A Geometric Path-Planning Algorithm in Cluttered Planar Environments Using Convex Hulls

2018 ◽  
Author(s):  
Nafiseh Masoudi ◽  
Georges Fadel
Keyword(s):  
Path Planning ◽  
Optimal Path ◽  
Visibility Graph ◽  
Convex Hulls ◽  
Planning Problem ◽  
Wire Routing ◽  
Visibility Algorithm ◽  
Planning Algorithm ◽  
Global Optimal ◽  
Path Planning Algorithm

The problem of finding a collision free path in an environment occupied by obstacles, known as path planning, has many applications in design of complex systems such as wire routing in automobile assemblies or motion planning for robots. Developing the visibility graph of the workspace is among the first techniques to address the path-planning problem. The visibility algorithm is efficient in finding the global optimal path. However, it is computationally expensive as it explores the entire workspace of the problem to create all non-intersecting segments of the graph. In this paper, we propose an algorithm based on the notion of convex hulls to generate the partial visibility graph from a given start point to a goal point in a 2D workspace cluttered with a number of disjoint polygonal convex or concave obstacles. The algorithm facilitates the attainment of the shortest path in a planar workspace while reducing the size of the visibility graph to explore.

scholarly journals Autonomous Vehicle Path Planning using Q-Learning

2021 ◽  
Vol 2128 (1) ◽  
pp. 012018
Author(s):  
Mohammed M S Ibrahim ◽  
Mostafa Rostom Atia ◽  
MW Fakhr
Keyword(s):  
Path Planning ◽  
Autonomous Vehicle ◽  
Optimal Path ◽  
Lookup Table ◽  
Planning Problem ◽  
Training Time ◽  
Q Learning ◽  
Starting Position ◽  
Planning Algorithm ◽  
Path Planning Problem

Abstract Path planning is vital in autonomous vehicle technology, from robots to self-driving cars and driverless trucks, it is impossible to navigate without a proper path planning algorithm, various algorithms exist Q-learning being one of them. Q-learning is used extensively in discrete applications as it is effective in finding solutions to these problems. This research investigates the possibility of using Q-learning for solving the local path planning problem with obstacle avoidance. Q-learning is split into two phases, the first being the training phase, and the second being the application phase. During training, Q-learning requires exponentially increasing training time based on the system’s state space. However, when Q-learning is applied it becomes as simple as a lookup table which allows it to run on even the simplest microcontrollers. Two simulations are conducted with varying environments. One to showcase the ability to learn the optimal path, the other to showcase the ability for learning navigation in variable environments. The first simulation was run on a static environment with one obstacle, with enough training episodes, Q-learning could solve the path planning problem with minimal movement steps. The second simulation focuses on a randomized environment, obstacles and the agent’s starting position are randomly chosen at the start of every episode. During testing, Q-learning was able to find a path to the target when a path did exist, as It was possible in certain configurations for the vehicle to be stuck in between obstacles with no feasible path or solution.

Sign in / Sign up

Export Citation Format

Share Document