Control methods of mobile robot rough-terrain trajectory tracking

Yang, Yi; Fu, Mengyin; Zhu, Hongtao; Xiong, Guangming; Chang-sheng, Sun

doi:10.1109/icca.2010.5524376

Cited by 8 publications

(4 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 8 displays the samples of trajectories which are passed by the robot using evaluated controllers in Test environment. 2 As this figure shows, the performance of RDL in following the given trajectory curves is the best. RDL never misses the wall and it avoids holes and pits more precisely because it utilizes the histories of features and previous steering commands which represent previous environment states.…”

Section: Resultsmentioning

confidence: 79%

“…Perception, planner, and motion control are conventional processes usually applied in robot navigation. 1,2 The most related researches have represented the robot environment as a grid map to determine the traversability of the outdoor terrains. These maps are created using some devices such as laser scanner, stereo camera, and Kinect camera.…”

Section: Introductionmentioning

confidence: 99%

“…So, it is more rational to reduce the input size by extracting proper features from these data and the controller architecture utilizes them. Some researchers have applied image preprocessing methods, for example, using a CCD color camera, to control the position of a goal-seeking robot, 9 estimating the 3D trajectory in unknown outdoor environments by a trinocular stereo camera, 10 applying both laser range finder and stereo vision to detect trajectory and designing the steering control law based on the kinematic equations of motion, 2,11 utilizing a stereo and mono vision to create a reinforcement learning (RL) approach for a wall follower robot to learn reactive behaviors. 12 Preprocessing techniques require complex mathematical computation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Combination of Recurrent Neural Network and Deep Learning for Robot Navigation Task in Off-Road Environment

2019

View full text Add to dashboard Cite

SUMMARYThis paper tackles the challenge of the necessity of using the sequence of past environment states as the controller’s inputs in a vision-based robot navigation task. In this task, a robot has to follow a given trajectory without falling in pits and missing its balance in uneven terrain, when the only sensory input is the raw image captured by a camera. The robot should distinguish big pits from small holes to decide between avoiding and passing over. In non-Markov processes such as the abovementioned task, the decision is done using past sensory data to ensure admissible performance. Applying images as sensory inputs naturally causes the curse of dimensionality difficulty. On the other hand, using sequences of past images intensifies this difficulty. In this paper, a new framework called recurrent deep learning (RDL) with combination of deep learning (DL) and recurrent neural network is proposed to cope with the above challenge. At first, the proper features are extracted from the raw image using DL. Then, these represented features plus some expert-defined features are used as the inputs of a fully connected recurrent network (as target network) to generate command control of the robot. To evaluate the proposed RDL framework, some experiments are established on WEBOTS and MATLAB co-simulation platform. The simulation results demonstrate the proposed framework outperforms the conventional controller based on DL for the navigation task in the uneven terrains.

show abstract

Section: Resultsmentioning

confidence: 79%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combination of Recurrent Neural Network and Deep Learning for Robot Navigation Task in Off-Road Environment

2019

View full text Add to dashboard Cite

show abstract

“…In the first step, the UGV's/ATV's kinematic constraints and surface geometry are used to plan a possible path toward the target position. Second, the UGV/ATV must track a selected path while considering the terrain's interactions with it [129].…”

Section: Navigation Plannersmentioning

confidence: 99%

A Comprehensive Overview of Control Algorithms, Sensors, Actuators, and Communication Tools of Autonomous All-Terrain Vehicles in Agriculture

Etezadi,

Eshkabilov

2024

Agriculture

View full text Add to dashboard Cite

This review paper discusses the development trends of agricultural autonomous all-terrain vehicles (AATVs) from four cornerstones, such as (1) control strategy and algorithms, (2) sensors, (3) data communication tools and systems, and (4) controllers and actuators, based on 221 papers published in peer-reviewed journals for 1960–2023. The paper highlights a comparative analysis of commonly employed control methods and algorithms by highlighting their advantages and disadvantages. It gives comparative analyses of sensors, data communication tools, actuators, and hardware-embedded controllers. In recent years, many novel developments in AATVs have been made due to advancements in wireless and remote communication, high-speed data processors, sensors, computer vision, and broader applications of AI tools. Technical advancements in fully autonomous control of AATVs remain limited, requiring research into accurate estimation of terrain mechanics, identifying uncertainties, and making fast and accurate decisions, as well as utilizing wireless communication and edge cloud computing. Furthermore, most of the developments are at the research level and have many practical limitations due to terrain and weather conditions.

show abstract

A new feature selection method based on task environments for controlling robots

Harandi

Derhami

Jamshidi

2019

Applied Soft Computing

View full text Add to dashboard Cite

Control methods of mobile robot rough-terrain trajectory tracking

Cited by 8 publications

References 8 publications

Combination of Recurrent Neural Network and Deep Learning for Robot Navigation Task in Off-Road Environment

Combination of Recurrent Neural Network and Deep Learning for Robot Navigation Task in Off-Road Environment

A Comprehensive Overview of Control Algorithms, Sensors, Actuators, and Communication Tools of Autonomous All-Terrain Vehicles in Agriculture

A new feature selection method based on task environments for controlling robots

Contact Info

Product

Resources

About