A Global Path Planning Algorithm for Robots Using Reinforcement Learning

Gao, Penggang; Liu, Zihan; Wu, Zongkai; Wang, Donglin

doi:10.1109/robio49542.2019.8961753

Cited by 30 publications

(15 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Sedangkan pada A* akan melakukan pencarian dengan memilih node baru berdasarkan biaya ke tujuan [5]. Proses path planning BFS dan Rapidly-exploring Random Trees (RRT) menghasilkan panjang jalur yang berbeda pada tiap lingkungan, pada empat scenario BFS menghasilkan jarak lebih pendek dan pada dua scenario RRT yang lebih pendek [13]. Pada percobaan BFS dan RRT tidak membandingkan jumlah node yang dibutuhkan [13].…”

Section: B Tinjauan State Of Artunclassified

“…Proses path planning BFS dan Rapidly-exploring Random Trees (RRT) menghasilkan panjang jalur yang berbeda pada tiap lingkungan, pada empat scenario BFS menghasilkan jarak lebih pendek dan pada dua scenario RRT yang lebih pendek [13]. Pada percobaan BFS dan RRT tidak membandingkan jumlah node yang dibutuhkan [13]. Pada pengujian algoritma A* menghasilkan waktu yang lebih cepat dibandingkan dengan algoritma basic theta* dan phi* dalam path planning, tetapi algoritma A* menghasilkan jarak jalur yang paling jauh [9].…”

Section: B Tinjauan State Of Artunclassified

See 1 more Smart Citation

Analisis Perbandingan Algoritma Breadth First Search (BFS) dan Algoritma A*

Aditiya¹,

Herdiana²

2021

Jurnal Ilmiah Telekomunikasi, Kendali dan Elektronika Terapan

View full text Add to dashboard Cite

Pada path planning terdapat beberapa metode yang dapat digunakan. Metode yang dapat digunakan untuk path planning yaitu algoritma Breadth First Search (BFS) dan algoritma A*. Metode yang digunakan tergantung dari keadaan lingkungan. Melakukan path planning pada tiap lingkungan dengan metode berbeda sehingga dapat mengetahui bagaimana peformansinya tiap metode tersebut. Sehingga tujuan penelitian ini adalah untuk membandingkan antara algoritma BFS dengan algoritma A* dalam melakukan path planning. Proses perbandingan dilakukan pada lima lingkungan pengujian. Lingkungan pengujian yang digunakan yaitu tanpa obstacle, obstacle trap, obstacle sederhana, obstacle maze dan obstacle narrow. Perbandingan algoritma BFS dan algoritma A* berdasarkan waktu eksekusi dan jumlah node yang dibutuhkan untuk melakukan path planning. Hasil penelitian menunjukkan bahwa pencarian jalur dari titik start ke titik goal dapat diselesaikan dengan algoritma BFS dan A*. Pada algoritma BFS maupun A* menghasilkan biaya jalur yang sama. Perbedaan terjadi pada node-node yang dibutuhkan algoritma BFS dan A* untuk menghasilkan jalur dari titik start hingga titik goal. Algoritma BFS membutuhkan node lebih banyak dibandingkan dengan A* untuk mencapai titik goal. Perbedaan jumlah node tersebut membuat waktu eksekusi menjadi berbeda. Waktu eksekusi pada algoritma BFS membutuhkan waktu lebih banyak dibandingkan dengan A*. Berdasarkan pengujian yang telah dilakukan maka algoritma A* lebih cepat dalam melakukan path planning. Tetapi pada lingkungan pengujian maze terjadi perbedaan waktu yang sedikit. Pada BFS memerlukan waktu tercepat 4,09 detik serta pada A* memerlukan waktu tercepat 3,88 detik. Serta pada lingkungan maze memiliki perbedaan jumlah node cukup sedikit yaitu 26 node. Hal tersebut membuktikan bahwa A* tidak selalu unggul jauh dengan BFS.

show abstract

Section: B Tinjauan State Of Artunclassified

Analisis Perbandingan Algoritma Breadth First Search (BFS) dan Algoritma A*

Aditiya¹,

Herdiana²

2021

Jurnal Ilmiah Telekomunikasi, Kendali dan Elektronika Terapan

View full text Add to dashboard Cite

show abstract

“…The authors modeled the problem as a sequential decision process and proposed a solution based on the Q-learning algorithm. In another work [20], the authors propose a path planning algorithm for robot navigation while combining global planning and reinforcement learning. First, a path graph is built using an accepted-rejected sampling where points are generated randomly and deleted if located in obstacle regions.…”

Section: Related Workmentioning

confidence: 99%

A generic framework for monitoring pollution plumes in emergencies using UAVs

Mokhtari

Bechkit

Rivano

2021

2021 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Monitoring air pollution plumes in emergency situations (industrial accidents, natural disasters, deliberate terrorist releases, etc.) becomes an issue of utmost importance in our society given the dramatic effects that the released pollutants can cause. Considering these situations, the pollution plume is strongly dynamic leading to a fast dispersion of pollutants in the atmosphere. Thus, the need for real-time response is very strong and a solution to get precise mapping of pollution dispersion is required to mitigate risks. However, monitoring and forecasting air quality in real time in such situations remains a highly challenging endeavour. In this paper, we suggest a systemic approach for monitoring dynamic air pollution based on aerial sensing (sensors mounted on UAVs). The proposed framework consists of a cycle with feedback loop which will constantly combine a spatio-temporal forecasting model based on a convolutional long short term memory (ConvLSTM) network with a data assimilation technique to get accurate pollution maps, while adjusting at each time the trajectories of drones following uncertainty forecasts. Our solution was evaluated and validated using a highly dynamic real world data set namely Fusion Field Trial 2007 (FFT07). The proposed strategy, together with the obtained evaluation results, are presented, and carefully analyzed.

show abstract

“…The authors claim that this method provides a more straightforward solution for robots with non-holonomic drive constraints. The paper [ 44 ] proposes an alternative for the existing best first search (BFS) and rapidly-exploring random trees algorithms, as they generate paths that are sharp and are not suitable for smooth navigation. Thus, the authors utilized the reinforcement learning-based Q-learning approach to automatically remove the paths that collide with the obstacles and thus produce smoother paths for seamless navigation.…”

Section: Introductionmentioning

confidence: 99%

Reinforcement Learning-Based Complete Area Coverage Path Planning for a Modified hTrihex Robot

Apuroop

Elara

et al. 2021

Sensors

View full text Add to dashboard Cite

One of the essential attributes of a cleaning robot is to achieve complete area coverage. Current commercial indoor cleaning robots have fixed morphology and are restricted to clean only specific areas in a house. The results of maximum area coverage are sub-optimal in this case. Tiling robots are innovative solutions for such a coverage problem. These new kinds of robots can be deployed in the cases of cleaning, painting, maintenance, and inspection, which require complete area coverage. Tiling robots’ objective is to cover the entire area by reconfiguring to different shapes as per the area requirements. In this context, it is vital to have a framework that enables the robot to maximize the area coverage while minimizing energy consumption. That means it is necessary for the robot to cover the maximum area with the least number of shape reconfigurations possible. The current paper proposes a complete area coverage planning module for the modified hTrihex, a honeycomb-shaped tiling robot, based on the deep reinforcement learning technique. This framework simultaneously generates the tiling shapes and the trajectory with minimum overall cost. In this regard, a convolutional neural network (CNN) with long short term memory (LSTM) layer was trained using the actor-critic experience replay (ACER) reinforcement learning algorithm. The simulation results obtained from the current implementation were compared against the results that were generated through traditional tiling theory models that included zigzag, spiral, and greedy search schemes. The model presented in the current paper was also compared against other methods where this problem was considered as a traveling salesman problem (TSP) solved through genetic algorithm (GA) and ant colony optimization (ACO) approaches. Our proposed scheme generates a path with a minimized cost at a lesser time.

show abstract

A Global Path Planning Algorithm for Robots Using Reinforcement Learning

Cited by 30 publications

References 19 publications

Analisis Perbandingan Algoritma Breadth First Search (BFS) dan Algoritma A*

Analisis Perbandingan Algoritma Breadth First Search (BFS) dan Algoritma A*

A generic framework for monitoring pollution plumes in emergencies using UAVs

Reinforcement Learning-Based Complete Area Coverage Path Planning for a Modified hTrihex Robot

Contact Info

Product

Resources

About