scholarly journals Assessing operator strategies for real-time replanning of multiple unmanned vehicles

2012 ◽  
Vol 6 (3) ◽  
pp. 221-231 ◽  
Author(s):  
Andrew S. Clare ◽  
Pierre C.P. Maere ◽  
M.L. Cummings
Keyword(s):  
Real Time ◽  
Unmanned Vehicles

Real-Time Path-Generation and Path-Following Using an Interoperable Multi-Agent Framework

2018 ◽  
Vol 06 (04) ◽  
pp. 231-250 ◽  
Author(s):  
Willson Amalraj Arokiasami ◽  
Prahlad Vadakkepat ◽  
Kay Chen Tan ◽  
Dipti Srinivasan
Keyword(s):  
Path Planning ◽  
Real Time ◽  
Collision Avoidance ◽  
Tracking Error ◽  
Unmanned Vehicles ◽  
Path Generation ◽  
Ground Vehicles ◽  
Agent Architectures ◽  
Time Path ◽  
Multi Agent

Autonomous unmanned vehicles are preferable in patrolling, surveillance and, search and rescue missions. Multi-agent architectures are commonly used for autonomous control of unmanned vehicles. Existing multi-robot architectures for unmanned aerial and ground robots are generally mission and platform oriented. Collision avoidance, path-planning and tracking are some of the fundamental requirements for autonomous operation of unmanned robots. Though aerial and ground vehicles operate differently, the algorithms for obstacle avoidance, path-planning and path-tracking can be generalized. Service Oriented Interoperable Framework for Robot Autonomy (SOIFRA) proposed in this work is an interoperable multi-agent framework focused on generalizing platform independent algorithms for unmanned aerial and ground vehicles. SOIFRA is behavior-based, modular and interoperable across unmanned aerial and ground vehicles. SOIFRA provides collision avoidance, and, path-planning and tracking behaviors for unmanned aerial and ground vehicles. Vector Directed Path-Generation and Tracking (VDPGT), a vector-based algorithm for real-time path-generation and tracking, is proposed in this work. VDPGT dynamically adopts the shortest path to the destination while minimizing the tracking error. Collision avoidance is performed utilizing Hough transform, Canny contour, Lucas–Kanade sparse optical flow algorithm and expansion of object-based time-to-contact estimation. Simulation and experimental results from Turtlebot and AR Drone show that VDPGT can dynamically generate and track paths, and SOIFRA is interoperable across multiple robotic platforms.

scholarly journals JAUS to EtherCAT Bridge: Toward Real-Time and Deterministic Joint Architecture for Unmanned Systems

2014 ◽  
Vol 2014 ◽  
pp. 1-20 ◽  
Author(s):  
Jie Sheng ◽  
Sam Chung ◽  
Leo Hansel ◽  
Don McLane ◽  
Joel Morrah ◽  
...  
Keyword(s):  
Real Time ◽  
Quantitative Data ◽  
Communication Systems ◽  
High Speed ◽  
Poor Performance ◽  
Unmanned Vehicles ◽  
Internal Communication ◽  
Area Network ◽  
Unmanned Systems ◽  
Level 3

The Joint Architecture for Unmanned Systems (JAUS) is a communication standard that allows for interoperability between Unmanned Vehicles (UVs). Current research indicates that JAUS-compliant systems do not meet real-time performance guidelines necessary for internal systems in UVs. However, there is a lack of quantitative data illustrating the performance shortcomings of JAUS or clear explanations on what causes these performance issues or comparisons with existing internal communication systems. In this research, we first develop a basic C++ implementation of JAUS and evaluate its performance with quantitative data and compare the results with published performance data of Controller Area Network (CAN) to determine the feasibility of the JAUS standard. Our results indicate that the main reason of JAUS’s poor performance lies in the latency inherent in the hierarchical structure of JAUS and the overhead of User Datagram Protocol (UDP) messages, which has been used with JAUS and is slower than the high-speed CAN. Additionally, UDP has no scheduling mechanism, which makes it virtually impossible to guarantee messages meeting their deadlines. Considering the slow and nondeterministic JAUS communication from subsystems to components, which is JAUS Level 3 compliance, we then propose a solution by bringing Ethernet for Control Automation Technology (EtherCAT) to add speed, deterministic feature, and security. The JAUS-EtherCAT mapping, which we called a JEBridge, is implemented into nodes and components. Both quantitative and qualitative results are provided to show that JEBridge and JAUS Level 3 compliance can bring not only interoperability but also reasonable performance to UVs.

scholarly journals REAL-TIME DEEP NEURAL NETWORKS FOR MULTIPLE OBJECT TRACKING AND SEGMENTATION ON MONOCULAR VIDEO

2021 ◽  
Vol XLIV-2/W1-2021 ◽  
pp. 15-20
Author(s):  
I. Basharov ◽  
D. Yudin
Keyword(s):  
Neural Networks ◽  
Real Time ◽  
Deep Neural Networks ◽  
Unmanned Vehicles ◽  
Geometric Constraints ◽  
Found Objects ◽  
Cuda Technology ◽  
Monocular Video ◽  
Found Object ◽  
Instance Segmentation

Abstract. The paper is devoted to the task of multiple objects tracking and segmentation on monocular video, which was obtained by the camera of unmanned ground vehicle. The authors investigate various architectures of deep neural networks for this task solution. Special attention is paid to deep models providing inference in real time. The authors proposed an approach based on combining the modern SOLOv2 instance segmentation model, a neural network model for embedding generation for each found object, and a modified Hungarian tracking algorithm. The Hungarian algorithm was modified taking into account the geometric constraints on the positions of the found objects on the sequence of images. The investigated solution is a development and improvement of the state-of-the-art PointTrack method. The effectiveness of the proposed approach is demonstrated quantitatively and qualitatively on the popular KITTI MOTS dataset collected using the cameras of a driverless car. The software implementation of the approach was carried out. The acceleration of the procedure for the formation of a two-dimensional point cloud in the found image segment was done using the NVidia CUDA technology. At the same time, the proposed instance segmentation module provides a mean processing time of one image of 68 ms, the embedding and tracking module of 24 ms using the NVidia Tesla V100 GPU. This indicates that the proposed solution is promising for on-board computer vision systems for both unmanned vehicles and various robotic platforms.

Design, modeling, and verification of a test bench for braking simulation of 1/4 vehicle

2019 ◽  
Vol 234 (5) ◽  
pp. 1425-1441
Author(s):  
Zeng-Cheng Liao ◽  
Xian-Xu ‘Frank’ Bai ◽  
Yang Li ◽  
Xue-Cai Deng ◽  
Jun Sun
Keyword(s):  
Kinetic Energy ◽  
Real Time ◽  
Test Bench ◽  
Processing System ◽  
Unmanned Vehicles ◽  
Signal Acquisition ◽  
Magnetic Powder ◽  
Road Surfaces ◽  
Signal Acquisition And Processing ◽  
Testing And Evaluation

Brake-by-wire systems are one of the key components in intelligent/unmanned vehicles that have attracted worldwide attention. Testing and evaluation of brake-by-wire systems are a significant step during the development of the technology of vehicular braking and further the advancement of intelligent/unmanned vehicles. Using the test bench to simulate different road adhesion coefficients (i.e. road surfaces) and to complete the testing and evaluation of the vehicle braking systems is of great significance and importance. A test bench for simulation of vehicular braking of 1/4 vehicle is presented and investigated in this article. It is composed of a motor, two rollers, a 1/4 vehicle suspension system, a magnetic powder clutch, a flywheel, sensors, a signal acquisition and processing system, and a controller. The wheel and vehicle speeds are simulated by the rollers and flywheel speeds, respectively. The translational kinetic energy of 1/4 vehicle is simulated by the rotational kinetic energy of the flywheel. The signal acquisition and processing system is used to acquire and process the experimental signals, such as rotational speeds and torques during tests. The magnetic powder clutch with adjusted applied currents in the test bench is used to real-timely simulate roads with different adhesion coefficients. Based on the working principle of the test bench and the fundamentals of vehicle dynamics, the prototype of the test bench is established and the simulation approach of the translational kinetic energy of 1/4 vehicle is investigated. The mathematical model of the real-time simulation about roads with different adhesion coefficients based on the magnetic powder clutch with real-time controllable transmitted torque is established. With the built 1/4 vehicle braking systems based on the test bench and road, and the corresponding established models, the comparison and analysis of the simulation results of various road surfaces are conducted. Experiments are sequentially implemented to verify the feasibility and effectiveness of the test bench.

scholarly journals Toward Real-time Assessment of Workload: A Bayesian Inference Approach

2019 ◽  
Vol 63 (1) ◽  
pp. 196-200 ◽  
Author(s):  
Ruikun Luo ◽  
Yifan Wang ◽  
Yifan Weng ◽  
Victor Paul ◽  
Mark J. Brudnak ◽  
...  
Keyword(s):  
Bayesian Inference ◽  
Real Time ◽  
High Mobility ◽  
Well Being ◽  
Unmanned Vehicles ◽  
Workload Management ◽  
Human In The Loop ◽  
High Workload ◽  
Human Operators

Workload management is of critical concern in teleoperation of unmanned vehicles, because high workload can lead to sub-optimal task performance and can harm human operators’ long-term well-being. In the present study, we conducted a human-in-the-loop experiment, where the human operator teleoperated a simulated High Mobility Multipurpose Wheeled Vehicle (HMMWV) and performed a secondary visual search task. We measured participants’ gaze trajectory and pupil size, based on which their workload level was estimated. We proposed and tested a Bayesian inference (BI) model for assessing workload in real time. Results show that the BI model can achieve an encouraging 0.69 F1 score, 0.70 precision, and 0.69 recall.

Sign in / Sign up

Export Citation Format

Share Document