Obstacle Distance Research Articles

A Deep Learning and Depth Image based Obstacle Detection and Distance Measurement Method for Substation Patrol Robot

Recently, substation patrol robot is gradually used to replace the manual inspection in order to improve the inspection efficiency as well as security and automation level of substation maintenance. The research of obstacle avoidance is a hot spot in substation intelligent patrol robot area. The emerging new generation of artificial intelligence (AI) technology provides a new way to solve the obstacle detection and distance measurement problem. To realize accurate, effective and real-time response to the environmental changes, a novel obstacle avoidance method based on deep learning and depth image is proposed. The core of this method is pixel-level instance segmentation between obstacles and roads, along with a pixel-level matching of obstacles’ segmentation mask and depth data. The effectiveness of the proposed method is validated by actual tests in real substation environment.

Experimental Investigation of Porosity, Installation Angle, Thickness and Second Layer of Permeable Obstacles on Density Current

This study explored the effect of porosity and installation angle, thickness (dimension) and second layer of permeable obstacles on density current control and trapping in the laboratory. For this purpose, an insoluble suspended polymer and two types of groove and cavity obstacles made from plexiglass sheets were selected. The experiments were conducted with two different concentrations, five different porosities, four different angles, four different thicknesses and two obstacle layers. The results showed that the optimum porosities for cavity and groove obstacles were 22 and 19%, respectively. In all experiments, the cavity trapping rates of 0.13% and 0.14% at 10% and 20% concentrations were higher than those of groove trapping. In addition, by increasing the angle, the rate of trapping decreased and its value was observed in the groove with the correlation coefficients of 0.995 and 0.981 compared to the cavity. The major effect of obstacles was found to be the flow deceleration where the average velocity in the cavity was obtained 3.62% higher than that in the groove. For the increased thickness with 10% porosity and groove type, the passage of materials from the obstacle further increased. By creating the second layer of obstacle, the passage of materials from the obstacle in the both groove and cavity increased, and the optimal distance of the second obstacle was 2.25 m from the first one.

An Indigenous Smart Cane System For The Visually Impaired With Sound Clap Location Ability

Guiding the visually impaired persons is always very tasking. Smart Canes previously designed for the blinds are relatively costly. Also, most of the available Smart Cane Systems used only remote control and buzzing methods for locating the cane if misplaced. However, no research work has handled how to locate the cane should the remote control itself be misplaced. This work designed an Indigenous Smart Cane System for the Visually Impaired with Sound Clap Location Ability. It is relatively less costly, easy to learn, and can be located by a natural means if misplaced. The system uses one ATMEGA328P Microcontroller which is programmed to control both input and output signals. It was designed with three low-cost 0.3m resolution HC-SR04 Ultrasonic Sensors to detect obstacles within the range of 2cm – 400cm in the front, left and right directions of the user and sends the current obstacle distance signal to the controller for processing. Receiving this signal, the controller determines which of the output devices (Piezo Speaker, Earphone and Vibrator Motor) to use and communicate the object distance to the user. In case the cane is misplaced, the user makes a sound clap which triggers the Sound Sensor to send signal to the controller for an audio feedback. This audio feedback is done by the Piezo Speaker. The system was programmed to use a very simple object – detection algorithm to help the user learn how to use the cane easily. This system has been tested and found to be relatively less costly, easy to learn and can be located with a clap sound.
 Keywords: Sound-Clap Location Ability, Visually Impaired, Indigenous Smart Cane, Sensors, Audio Feedback; Object Distance; Microcontroller.

Effects of unequal blockage ratio and obstacle spacing on wave speed and overpressure during flame propagation in stoichiometric H2/O2

Experiments on flame propagation and detonation onset behind two solid obstructions were carried out in premixed stoichiometric hydrogen–oxygen mixtures at 20 kPa in a closed-ended tube. Obstacles with three different blockage ratios (25%, 40%, and 80%) were used, and the arrangement between the obstacles was changed in terms of blockage distribution (increasing, decreasing, and equivalent); obstacle distance (38, 76, and 114 mm); and opening geometry. Changes in the obstacle pair characteristics resulted in shocks with distinct intensities and averaged Mach numbers, $$ \overline{{M_{\text{s}} }} $$ , that propagated into the undisturbed mixture; $$ \overline{{M_{\text{s}} }} $$ varied from 1.3 to 4.5 for the range of obstacle pairs tested. Four distinct deflagration-to-detonation transition (DDT) timescale groups were identified with average values varying between 0.7 and 12 ms. Obstacle pairs with increasing blockage ratio (25–80% and 40–80%) resulted in jet ignition downstream of the second obstacle, reducing the DDT length and timescale significantly when compared to the other obstruction combinations investigated. Soot foil records showed that detonation onset was via ignition from one or more hot spots near the flame front for all obstacle pairs that resulted in jet ignition. The results from this work suggest that channels with irregular obstacle patterns can experience faster detonation onset depending on the position of the obstructions relative to the ignition point.

Pedestrian evacuation simulation in the presence of an obstacle using self-propelled spherocylinders

We explore the role that the obstacle position plays in the evacuation time of agents when leaving a room. To this end, we simulate a system of nonsymmetric spherocylinders that have a prescribed desired velocity and angular orientation. In this way, we reproduce the nonmonotonous dependence of the pedestrian flow rate on the obstacle distance to the door. For short distances, the obstacle delays the evacuation because the exit size is effectively reduced; i.e., the distance between the obstacle and the wall is smaller than the door width. By increasing the obstacle distance to the door, clogging is reduced leading to an optimal obstacle position (maximum flow rate) in agreement with results reported in numerical simulations of pedestrian evacuations and granular flows. For further locations, however, a counterintuitive behavior occurs as the flow rate values fall again below the one corresponding to the case without obstacle. Analyzing the head-times distribution, we evidence that this new feature is not linked to the formation of clogs, but is caused by a reduction of the efficiency of the agent's instantaneous flow rate when the exit is not blocked.

Navigational Analysis of Multiple Humanoids Using a Hybridized Rule Base-Sugeno Fuzzy Controller

This paper describes a rule base-Sugeno fuzzy hybrid controller for path planning of single as well as multiple humanoid robots in cluttered environments. Initially, sensor outputs regarding the obstacle distances are used as inputs to the rule base model, and turning angle is obtained as the output. The rule-based analysis is used for training the fuzzy controller with membership functions. The output from the rule base model along with other regular inputs is supplied to a Sugeno fuzzy model, and effective turning angle is obtained as the final output to avoid the obstacles present in the environment and navigate the humanoids safely to their target points. The proposed hybrid controller is tested on a V-REP simulation platform, and the simulation results are validated in an experimental set-up. To avoid the possibility of any inter-collision during navigation of multiple humanoids on a common platform, a Petri-net scheme is integrated along with the proposed hybrid model. Finally, the results obtained from simulation and experimental platforms are compared against each other with proper agreement and minimal percentage of deviations. To validate the proposed controller, it has also been tested against another existing navigational approach, and satisfactory performance enhancement has been observed.

Blind people’s direction support system using ultrasonic, color sensor with fuzzy logic

Someone who has impaired vision or blindness has a condition that is susceptible to dangers that are around him such as the presence of animals, vehicles or obstructions. People with visual impairments carry out their activities based on estimates of their feelings. One obstacle that is often faced by blind people is that if they are in a new environment, they must recognize patterns around the environment such as roads, rooms, or certain important places.his research in the form of blind shoe design using ultrasonic, color sensor and Arduino Uno microcontroller, with fuzzy logic, aims to help blind people, especially children who are educated in extraordinary schools so that in carrying out their learning activities do not experience obstacles in recognizing the school environment, such as study rooms or classrooms, toilets and roads that are in the school environment. The method used is experimental, the color sensor can read input in the form of red, blue, green and yellow which is the blind pointing pathway, and ultrasonic can read the distance of obstacles in front of the shoe using an intelligent fuzzy logic system. A pair of shoes uses Bluetooth communication, output in the form of sound through earphones as a guide to the classroom or toilet.

Obstacle distance measurement based on binocular vision for high-voltage transmission lines using a cable inspection robot.

Obstacle distance measurement is one of the key technologies for cable inspection robots on high-voltage transmission lines. This article develops a novel method based on binocular vision for extracting the feature points of images and reconstructing 3D scenes. The proposed method seamlessly incorporates camera calibration, dense stereo matching, and 3D reconstruction. We apply a novel calibration method to acquire intrinsic and extrinsic parameters and use an improved Semi-Global Matching (SGM) algorithm based on the least squares fitting interpolation to refine the basic disparity map. Based on the depth information of the optimized disparity map and the principle of binocular vision measurement, a model is established to estimate the distance of an obstacle from the cable inspection robot. Extensive experiments show that the proposed method achieves an estimation accuracy of less than 5% from 0.5 m to 5.0 m, offering extremely high distance estimation accuracy and robustness. The study improves the autonomy and intelligence of inspection robots used in the power industry.

V-REP-based navigation of automated wheeled robot between obstacles using PSO-tuned feedforward neural network

Abstract This paper describes the navigation of an automated Pioneer P3-DX wheeled robot between obstacles using particle swarm optimization (PSO) algorithm tuned feedforward neural network (FNN). This PSO algorithm minimizes the mean square error between the actual and predicted values of the FNN. In this work, 2 separate DC motors and 16 ultrasonic sensors have been used for making differential drive steering angle and for collecting the distance from obstacles, respectively. The proposed without tuned FNN and PSO-tuned FNN receives obstacle's distance as inputs form ultrasonic sensors and control the steering angle of a differential drive of automated Pioneer P3-DX wheeled robot as output. We have compared the results between without tuned FNN and PSO-tuned FNN, and it has been found that PSO-tuned FNN gives a better trajectory and takes less distance to reach the target. Virtual Robot Experimentation Platform software has been used to design the real-time simulation results. A comparative study between without tuned FNN and PSO-tuned FNN verifies the effectiveness of PSO-tuned FNN for automated Pioneer P3-DX wheeled robot navigation. Also, we have compared this winner PSO-tuned FNN to the previously developed PSO-optimized Fuzzy Logic Controller navigational technique to show the authenticity and real-time implementation of PSO-tuned FNN.

Depth estimation for advancing intelligent transport systems based on self‐improving pyramid stereo network

In autonomous driving, stereo vision-based depth estimation technology can help to estimate the distance of obstacles accurately, which is crucial for correctly planning the path of the vehicle. Recent work has formulated the stereo depth estimation problem into a deep learning model with convolutional neural networks. However, these methods need a lot of post-processing and do not have strong adaptive capabilities to ill-posed regions or new scenes. In addition, due to the difficulty of the labelling the ground truth depth for real circumstance, training data for the system is limited. To overcome the above problems, the authors came up with self-improving pyramid stereo network, which can not only get a direct regression disparity without complicated post-processing but also be robust in ill-posed area. Moreover, by online learning, the proposed model can not only address the data limitation problem but also save the time spent on training and hardware resources in practice. At the same time, the proposed model has a self-improving ability to new scenes, which can quickly adjust the model according to the test data in time and improve the accuracy of prediction. Experiments on Scene Flow and KITTI data set demonstrate the effectiveness of the proposed network.

RETRACTED: Design of basketball robot based on behavior-based fuzzy control

Aiming at the strong dependence on environmental information in traditional algorithms, the path planning of basketball robots in an unknown environment, and improving the safety of autonomous navigation, this article proposes a path planning algorithm based on behavior-based module control. In this article, fuzzy control theory is applied to the behavior control structure, and these two path planning algorithms are combined to solve the path planning problem of basketball robots in an unknown environment. First, the data of each sensor of the basketball robot configuration are simply fused. Then, the obstacle distance parameters in the three directions of front, left, and right are simplified and fuzzified. Then combined with the target direction parameters, the speed, and steering of the basketball robot are controlled by fuzzy rule reasoning to realize path planning. The simulation results show that the basketball robot can overcome the uncertainty in the environment, effectively achieve good path planning, verify the feasibility of the fuzzy control algorithm, and demonstrate the validity and correctness of the path planning strategy.

On the Structure of an Impact Jet with Flow Swirling and Combustion

The PIV and PLIF OH techniques were used to experimentally study the structure of a jet impinging on a flat obstacle with swirling and combustion of a propane–air mixture with an equivalence ratio of 0.7 for a nozzle–obstacle distance H/d = 1 and 3 and in the absence of the obstacle. The swirl ratio was 0.41 and 1.0, and the Reynolds number was 5 000. It is concluded that for both values of the swirl ratio, the presence of the impact surface leads to the formation of an extended central cone-shaped recirculation zones. For H/d = 3 and 2, the OH fluorescence intensity near the impact surface and inside the recirculation zone is significantly reduced. This effect may be caused by a decrease in the temperature of combustion products in the recirculation zone, including in the vicinity of the flame front.

A Novel Micromixer with Fractal Obstacles Designed Based on Generalized Murray’s Law

AbstractIn this paper, we designed tree-like branch structures with two levels according to generalized Murray’s law, and set them as obstacles in microchannels. The influence of the mixing performance of the micromixer was studied by changing the geometry sizes of obstacles, the branch angles and the distances between obstacles. The mixing performance of micromixer was analyzed from three aspects, including mixing efficiency, flow behavior, and pressure drop.We designed three different obstacles A, B, and C, the micromixer with obstacle C has the highest mixing efficiency. For the distances of obstacles, the micromixer with smaller distance has the higher mixing efficiency. The micromixer with branch angle of 75° has the highest mixing efficiency in the range of Reynolds number (Re) 0.1 to 150.

An open-pit mine roadway obstacle warning method integrating the object detection and distance threshold model

In order to solve the problem that the current driving warning method cannot adapt to the unstructured road in open-pit mine, this paper proposes an early warning method that integrates target detection and obstacle distance threshold. Firstly, the original Mask R-CNN detection framework was improved according to the characteristics of open-pit mine obstacles, and dilated convolution was introduced into the framework network to expand the receptive field range without reducing the feature map to ensure the detection accuracy of larger targets. Then, a linear distance factor was constructed based on the target detection results to represent the depth information of obstacles in the input image, and an SVM warning model was established. Finally, in order to ensure the generalization ability of the warning model, transfer learning method was adopted to carry out pre-training of the network in COCO data set, and both the C5 stage and detection layer were trained in the data collected in the field. The experimental results show that the accuracy and recall of the proposed method reach 98.47% and 97.56% in the field data detection, respectively, and the manually designed linear distance factor has a good adaptability to the SVM warning model.

Navigational analysis of multiple humanoids using a hybrid regression-fuzzy logic control approach in complex terrains

In the current work, a hybrid navigational control architecture combining regression analysis with fuzzy logic control has been proposed for smooth and hassle-free motion planning of humanoids. In the proposed hybrid scheme, sensory information regarding obstacle distances are initially supplied to the regression controller, and an interim turning angle is obtained as the preliminary output based on the preloaded training pattern of the regression model. In the next phase, interim turning angle is again supplied to the fuzzy controller to generate the ultimate turning angle which eventually guides the humanoid to take a safe direction of turn while avoiding any obstacle present in the work environment. The working of the developed hybrid model is validated through simulation and real-time environments, and satisfactory results have been obtained from comparisons of selected navigational parameters along with a minimal percentage of deviations. To avoid possible chances of inter-collision for navigation of multiple humanoids in a common platform, a Petri-Net model has been integrated with the developed hybrid control scheme. Finally, the developed motion planning model is also assessed against another existing navigational controller, and significant performance enhancement is obtained.

Navigational analysis of a humanoid using genetic algorithm with vision assistance

In this paper, a novel vision assisted genetic algorithm based navigational controller has been designed for smooth and collision-free path generation of a humanoid robot. Here, sensory information regarding the nearest obstacle distance and path left to the destination are considered as the inputs to the genetic algorithm controller, and necessary turning angle is generated as the required output to avoid the obstacles present in the path and advance towards the destination. The vision based technique is integrated along with the sensor based navigational model to assist in deciding a safe direction of turn in case the humanoid encounters a dead end situation while negotiating with complicated obstacle settings. The developed model has been verified by navigational analysis of a NAO humanoid in a V-REP simulation arena. The simulation results are also validated against an experimental set-up prepared under laboratory conditions that resembles the simulation arena. The results obtained from both the platforms are compared in terms of selected navigational parameters, and a close agreement has been found between them with a minimal percentage of errors. Finally, the developed model is also evaluated against other existing navigational schemes, and substantial performance improvements have been observed.

Numerical Investigations of a Passive Micromixer Based on Minkowski Fractal Principle

Abstract This paper is mainly to study the mixing efficiency and pressure drop of the Minkowski fractal obstacle micromixers. The mixing efficiency of primary Minkowski fractal obstacle (PMFO) micromixer and secondary Minkowski fractal obstacle (SMFO) micromixer are compared at five kinds of Reynolds numbers. With the increase of obstacle height and the decrease of distance, the chaotic convection in the microchannel is enhanced. Especially at obstacle height (h) = 0.2 mm, obstacle distance (D) = 0.15 mm, and Re = 100, the vortex caused by the Minkowski fractal obstacle structure is more obvious. In addition, vortex phenomenon increases the contact area of two fluids and enhances chaotic convection. It shows that the flow direction of the fluid in the microchannel varies significantly.

Lattice Boltzmann simulation of droplet deformation and breakup due to collision with obstacles in microchannel

In this paper, passive droplet deformation and breakup in microchannel with obstacles were investigated. The lattice Boltzmann method based on pseudo-potential method was utilized for the two-phase flow simulation in two cases, symmetric and asymmetric deformation and breakup of droplet. Validation was performed by several well-known benchmarks. The effects of various flow and physical parameters such as Reynolds number, contact angle, channel height and droplet diameter in symmetric case were investigated. It was seen that in the symmetric case, by increasing triangular obstacles’ height, initial diameter of droplet and contact angle, the maximum velocity of droplet in narrow section increased. Also, the effects of obstacles distance and size on the droplet motion and breakup in asymmetric case were investigated.

Open Vision System for Low-Cost Robotics Education

Vision devices are currently one of the most widely used sensory elements in robots: commercial autonomous cars and vacuum cleaners, for example, have cameras. These vision devices can provide a great amount of information about robot surroundings. However, platforms for robotics education usually lack such devices, mainly because of the computing limitations of low cost processors. New educational platforms using Raspberry Pi are able to overcome this limitation while keeping costs low, but extracting information from the raw images is complex for children. This paper presents an open source vision system that simplifies the use of cameras in robotics education. It includes functions for the visual detection of complex objects and a visual memory that computes obstacle distances beyond the small field of view of regular cameras. The system was experimentally validated using the PiCam camera mounted on a pan unit on a Raspberry Pi-based robot. The performance and accuracy of the proposed vision system was studied and then used to solve two visual educational exercises: safe visual navigation with obstacle avoidance and person-following behavior.

A deep learning platooning-based video information-sharing Internet of Things framework for autonomous driving systems

To enhance the safety and stability of autonomous vehicles, we present a deep learning platooning-based video information-sharing Internet of Things framework in this study. The proposed Internet of Things framework incorporates concepts and mechanisms from several domains of computer science, such as computer vision, artificial intelligence, sensor technology, and communication technology. The information captured by camera, such as road edges, traffic lights, and zebra lines, is highlighted using computer vision. The semantics of highlighted information is recognized by artificial intelligence. Sensors provide information on the direction and distance of obstacles, as well as their speed and moving direction. The communication technology is applied to share the information among the vehicles. Since vehicles have high probability to encounter accidents in congested locations, the proposed system enables vehicles to perform self-positioning with other vehicles in a certain range to reinforce their safety and stability. The empirical evaluation shows the viability and efficacy of the proposed system in such situations. Moreover, the collision time is decreased considerably compared with that when using traditional systems.