Neural-network-based learning control for the high-speed path tracking of unmanned ground vehicles

2003 ◽  
Author(s):  
Xin Xu ◽  
Han-Gen He
Keyword(s):  
Neural Network ◽  
High Speed ◽  
Learning Control ◽  
Path Tracking ◽  
Unmanned Ground Vehicles ◽  
Ground Vehicles

High-speed navigation of unmanned ground vehicles on uneven terrain using potential fields

Robotica ◽  
2007 ◽  
Vol 25 (4) ◽  
pp. 409-424 ◽  
Author(s):  
Shingo Shimoda ◽  
Yoji Kuroda ◽  
Karl Iagnemma
Keyword(s):  
Potential Field ◽  
High Speed ◽  
Longitudinal Velocity ◽  
Unmanned Ground Vehicles ◽  
Performance Bounds ◽  
Ground Vehicles ◽  
Computationally Efficient ◽  
Field Methods ◽  
Local Maxima ◽  
Natural Terrain

SUMMARYMany applications require unmanned ground vehicles (UGVs) to travel at high speeds on sloped, natural terrain. In this paper, a potential field-based method is proposed for UGV navigation in such scenarios. In the proposed approach, a potential field is generated in the two-dimensional “trajectory space” of the UGV path curvature and longitudinal velocity. In contrast to traditional potential field methods, dynamic constraints and the effect of changing terrain conditions can be easily expressed in the proposed framework. A maneuver is chosen within a set of performance bounds, based on the local potential field gradient. It is shown that the proposed method is subject to local maxima problems, rather than local minima. A simple randomization technique is proposed to address this problem. Simulation and experimental results show that the proposed method can successfully navigate a small UGV between predefined waypoints at speeds up to 7.0 m/s, while avoiding static hazards. Further, vehicle curvature and velocity are controlled during vehicle motion to avoid rollover and excessive side slip. The method is computationally efficient, and thus suitable for onboard real-time implementation.

scholarly journals A Monocular Vision Obstacle Avoidance Method Applied to Indoor Tracking Robot

Drones ◽  
2021 ◽  
Vol 5 (4) ◽  
pp. 105
Author(s):  
Shubo Wang ◽  
Ling Wang ◽  
Xiongkui He ◽  
Yi Cao
Keyword(s):  
Neural Network ◽  
Obstacle Avoidance ◽  
Small Sample ◽  
Learning Rate ◽  
Monocular Vision ◽  
Unmanned Ground Vehicles ◽  
Indoor Environments ◽  
Ground Vehicles ◽  
Data Set ◽  
Control Command

The overall safety of a building can be effectively evaluated through regular inspection of the indoor walls by unmanned ground vehicles (UGVs). However, when the UGV performs line patrol inspections according to the specified path, it is easy to be affected by obstacles. This paper presents an obstacle avoidance strategy for unmanned ground vehicles in indoor environments. The proposed method is based on monocular vision. Through the obtained environmental information in front of the unmanned vehicle, the obstacle orientation is determined, and the moving direction and speed of the mobile robot are determined based on the neural network output and confidence. This paper also innovatively adopts the method of collecting indoor environment images based on camera array and realizes the automatic classification of data sets by arranging cameras with different directions and focal lengths. In the training of a transfer neural network, aiming at the problem that it is difficult to set the learning rate factor of the new layer, the improved bat algorithm is used to find the optimal learning rate factor on a small sample data set. The simulation results show that the accuracy can reach 94.84%. Single-frame evaluation and continuous obstacle avoidance evaluation are used to verify the effectiveness of the obstacle avoidance algorithm. The experimental results show that an unmanned wheeled robot with a bionic transfer-convolution neural network as the control command output can realize autonomous obstacle avoidance in complex indoor scenes.

Experimental research on the muzzle response characteristics of small unmanned ground vehicles with small arms

2021 ◽  
pp. 095440622110570
Author(s):  
Yugang Ding ◽  
Kedong Zhou ◽  
Lei He ◽  
Haomin Yang
Keyword(s):  
High Speed ◽  
Threshold Function ◽  
Measurement Unit ◽  
Unmanned Ground Vehicles ◽  
Vehicle Speed ◽  
Ground Vehicles ◽  
Small Arms ◽  
Inertial Measurement ◽  
Variable Threshold ◽  
Response Characteristics

The muzzle response is the main feature affecting the firing accuracy of weapons. To research the muzzle response characteristics of small unmanned ground vehicles with small arms (SUGVsSA) during shooting, this paper designs a test method that combines an inertial measurement system (IMS) with a high-speed photogrammetric system (HSPS) to measure the muzzle response. That is, an inertial measurement unit (IMU) is fixed onto the gun body to record the three-dimensional angular motion of the barrel; meanwhile, a high-speed camera is used to capture the characteristic markers of the unmanned ground vehicle from the side. After data processing, the muzzle response curves during four consecutive firings when the vehicle is running at different speeds and firing angles are obtained. Considering the presence of noise in muzzle response signals, the wavelet threshold de-noising (WTD) algorithm based on a novel variable threshold function is used to de-noise the test signal. The processing results demonstrate that the WTD algorithm based on the novel variable threshold function can not only suppress noise in the muzzle response signal but also retain the local details of the signal. The combination of the IMS and HSPS complements the muzzle response data and can comprehensively and accurately reflect the muzzle response characteristics of SUGVsSA. As the vehicle speed and firing angle increase, the muzzle vibration intensifies, only when the vehicle speed is 0.3 m/s, and the muzzle maximum elevation angle displacement after each firing decreases when it is stationary. The results presented in this paper may provide a workable reference for understanding the muzzle response characteristics of SUGVsSA and evaluating the firearm compatibility of other unmanned systems.

Research on Wheeled Robot Path Tracking Based on Fuzzy Neural Network

2014 ◽  
Vol 602-605 ◽  
pp. 1270-1274
Author(s):  
Shuo Yang ◽  
Bo Quan Zhang ◽  
Mao Ying Jia
Keyword(s):  
Neural Network ◽  
High Speed ◽  
Fuzzy Neural Network ◽  
Path Tracking ◽  
Second Stage ◽  
Wheeled Robot ◽  
Fuzzy Neural ◽  
Turning Radius ◽  
Path Tracking Method ◽  
Robot Path

To solve the problem that wheeled robot in the path tracking is prone to slip or roll over at a sharp curve, structure of the wheeled robot and its path tracking features are analyzed, and a new Fuzzy-Neural Network (FNN) based path tracking method of two-stage (route and speed) control is proposed. In the first stage, a FNN controller determines the robot’s turning radius by processing robot’s pose information. In the second stage, the secondary controller adjusts angular and linear velocities by taking advantage of the turning radius and condition of the path ahead. The experiments show that the controlled robot can track the planned path accurately and robustly when it runs at high speed; the process of path tracking is stable and no slipping and rolling occur.

scholarly journals Indoor Navigation of Unmanned Grounded Vehicle using CNN

2020 ◽  
Vol 8 (6) ◽  
pp. 1766-1771
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Software Architecture ◽  
Real Time ◽  
Indoor Environment ◽  
Indoor Navigation ◽  
Training Data ◽  
Unmanned Ground Vehicles ◽  
Ground Vehicles ◽  
Complete Process

This paper presents a hardware and software architecture for an indoor navigation of unmanned ground vehicles. It discusses the complete process of taking input from the camera to steering the vehicle in a desired direction. Images taken from a single front-facing camera are taken as input. We have prepared our own dataset of the indoor environment in order to generate data for training the network. For training, the images are mapped with steering directions, those are, left, right, forward or reverse. The pre-trained convolutional neural network(CNN) model then predicts the direction to steer in. The model then gives this output direction to the microprocessor, which in turn controls the motors to transverse in that direction. With minimum amount of training data and time taken for training, very accurate results were obtained, both in the simulation as well as actual hardware testing. After training, the model itself learned to stay within the boundary of the corridor and identify any immediate obstruction which might come up. The system operates at a speed of 2 fps. For training as well as making real time predictions, MacBook Air was used.

Sign in / Sign up

Export Citation Format

Share Document