FPGA-based autonomous parking of a car-like robot using Fuzzy Logic Control

Scicluna, Nicolanne; Gatt, Edward; Casha, Owen; Grech, Ivan; Micallef, Joseph

doi:10.1109/icecs.2012.6463759

Cited by 10 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Scicluna et al 10 suggested a hardware solution to the autonomous parking problem of a car-like robot. They designed a Fuzzy Logic Controller for parallel and vertical parking.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Deep reinforcement learning‐based autonomous parking design with neural network compute accelerators

Özeloğlu

Gürbüz

San

2021

Concurrency and Computation

View full text Add to dashboard Cite

We describe the design and implementation of an autonomous prototype vehicle which finds an empty parking slot in a parking area, and parks itself in the empty parking slot, using neural networks based on deep reinforcement learning (RL). To perform an autonomous parking procedure for our prototype vehicle, two different artificial neural networks (ANNs) are trained using a deep RL Algorithm in a simulation environment and embedded into the computing platform of the prototype car.One of the ANNs enables the vehicle to drive autonomously in the parking environment. At the same time, an image processing algorithm is used to determine whether a parking slot is empty. When the image processing algorithm finds a suitable parking slot, a different ANN is activated and performs a safe parking procedure. However, ANN-based machine learning techniques require high processing power and impose a high computational burden on embedded CPU and GPU platforms. To alleviate the computational burden, one can achieve higher performance and less power consumption using an application-specific hardware design, where logic resources are fully exploited according to the algorithm of interest, in an energy-efficient manner. In this article, hardware accelerators for our ANN models are designed and generated via the Vivado high-level synthesis (HLS) tool, targeting an ARM based programmable SoC platform, ZedBoard. Our ANN accelerators have achieved a speedup of 17x as compared to an ARM software implementation. For deeper fully-connected layers used in deep RL-based solutions, function-level parallelism (Vivado's dataflow) is employed to improve the computational efficiency. Our proposed stage-level description for fully connected layers outperforms recent studies in terms of computation time.

show abstract

“…Scicluna et al 10 suggested a hardware solution to the autonomous parking problem of a car-like robot. They designed a Fuzzy Logic Controller for parallel and vertical parking.…”

Section: Related Workmentioning

confidence: 99%

“…They designed a Fuzzy Logic Controller for parallel and vertical parking. The main difference of our study with the work by Scicluna et al 10 is that we utilized a solution based on ANNs which are specifically trained for self-parking tasks.…”

Section: Related Workmentioning

confidence: 99%

Deep reinforcement learning‐based autonomous parking design with neural network compute accelerators

Özeloğlu

Gürbüz

San

2021

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Penelitian menggunakan CLMR telah banyak dilakukan oleh beberapa peneliti seperti pada (Baturone et al 2004;Chiu et al 2005;El-Khatib & Hamilton, 2006;Gucrrcro et al 2014;Liu et al 2011;Ouadah et al 2006;Scicluna et al 2012;Abatari et al 2013). Pada makalah ini memaparkan suatu implementasi CLMR dengan Ackerman steering pada sistem robot penjejak garis.…”

Section: Pendahuluanunclassified

Kendali Logika Fuzzy pada Car Like Mobile Robot (CLMR) Penjejak Garis

et al. 2018

View full text Add to dashboard Cite

ABSTRAKMakalah ini memaparkan perancangan sistem kendali logika fuzzy untuk mengatur kecepatan dan arah sudut steering pada car like mobile robot (CLMR) dengan menggunakan metode Ackermann steering. CLMR penjejak garis dirancang menggunakan 16 buah photodiode, dan terdapat 7 buah membership fuzzfikasi dari pembacaan error dan last error sehingga terbentuk 49 aturan. Untuk menguji perfoma kendali fuzzy pada sistem CLMR dalam mengikuti lintasan garis maka dilakukan pengujian dengan bentuk lintasan berupa garis lurus dan berbelok serta zig-zag dalam satu lintasan putar. Proses variasi nilai keanggotaan fuzzifikasi masukan dan defuzzifikasi keluaran dilakukan sebanyak lima kali. Dari hasil pengujian diperoleh bahwa kendali logika fuzzy yang diaplikasikan pada sistem mampu membuat pergerakan CLMR sukses mengikuti lintasan uji selama 9,38 detik lebih baik 0,53 detik dari kendali PID. Selanjutnya, hasil rancangan sistem CLMR ini merupakan sebuah prototipe self-driving car.Kata kunci: car like mobile robot, robot penjejak garis, fuzzy, self-driving car ABSTRACTThis paper describes the design of a fuzzy logic control system to adjust the speed and direction of the angle of the steering on the car like mobile robot (CLMR) using the Ackermann steering method. CLMR line tracking is designed using 16 photodiode pieces, and there are 7 fuzzfication membership from reading error and last error so that 49 rules are formed. To test the fuzzy control performance on the CLMR system in following the line trajectory, it was tested with the form of a straight line and a turn and a zigzag in a rotary track. The process of varying input membership fuzzification values and output defuzzification is done five times. From the test results, it was found that the fuzzy logic control applied to the system was able to make CLMR movement successfully followed the test path for 9.38 seconds better than 0.53 seconds of PID control. Furthermore, the results of the CLMR system design are a prototype self-driving car.Keywords: car like mobile robot, line tracking robot, fuzzy, self-driving car

show abstract

“…This parking operation is useful for parallel parking operation. [7] In 2012, N.Scicluna,E.Gatt,O.Casha,I.Grech,andJ.Micallef proposed a paper for autonomous parking using fuzzy logic control in International conference named "Fpga-based autonomous parking of a car-like robot using fuzzy logic control" [8] After having knowledge about these algorithms, generally systems need parking space more than 1.4 times of the vehicle length for the parking process. [9] …”

Section: Previous Parking Algorithmsmentioning

confidence: 99%