Interval-based Solutions for Reliable and Safe Navigation of Intelligent Autonomous Vehicles

Comparative Analysis Between Recurrent Convolutional and Convolutional Neural Networks for Horizon Detection

Journal of Computational and Theoretical Nanoscience ◽

10.1166/jctn.2020.9078 ◽

2020 ◽

Vol 17 (9) ◽

pp. 4364-4367

Author(s):

Shreya Srinarasi ◽

Seema Jahagirdar ◽

Charan Renganathan ◽

H. Mallika

Keyword(s):

Neural Networks ◽

Convolutional Neural Networks ◽

Autonomous Vehicles ◽

Semantic Segmentation ◽

Unmanned Vehicles ◽

Obstacle Detection ◽

Ground Vehicles ◽

Safe Navigation ◽

Horizon Line ◽

Flight Parameters

The preliminary step in the navigation of Unmanned Vehicles is to detect and identify the horizon line. One method to locate the horizon and obstacles in an image is through a supervised learning, semantic segmentation algorithm using Neural Networks. Unmanned Aerial Vehicles (UAVs) are rapidly gaining prominence in military, commercial and civilian applications. For the safe navigation of UAVs, there poses a requirement for an accurate and efficient obstacle detection and avoidance. The position of the horizon and obstacles can also be used for adjusting flight parameters and estimating altitude. It can also be used for the navigation of Unmanned Ground Vehicles (UGV), by neglecting the part of the image above the horizon to reduce the processing time. Locating the horizon and identifying the various obstacles in an image can help in minimizing collisions and high costs due to failure of UAVs and UGVs. To achieve a robust and accurate system to aid navigation of autonomous vehicles, the efficiency and accuracy of Convolutional Neural Networks (CNN) and Recurrent-CNNs (RCNN) are analysed. It is observed via experimentation that the RCNN model classifies test images with higher accuracy.

Download Full-text

A Global Path Planner for Safe Navigation of Autonomous Vehicles in Uncertain Environments

Sensors ◽

10.3390/s20216103 ◽

2020 ◽

Vol 20 (21) ◽

pp. 6103

Author(s):

Mohammed Alharbi ◽

Hassan A. Karimi

Keyword(s):

Autonomous Vehicles ◽

Dynamic Environments ◽

Complex Dynamic ◽

Uncertain Environments ◽

Sensor Performance ◽

Safe Navigation ◽

Path Planner ◽

Main Barrier ◽

Safe Path

Autonomous vehicles (AVs) are considered an emerging technology revolution. Planning paths that are safe to drive on contributes greatly to expediting AV adoption. However, the main barrier to this adoption is navigation under sensor uncertainty, with the understanding that there is no perfect sensing solution for all driving environments. In this paper, we propose a global safe path planner that analyzes sensor uncertainty and determines optimal paths. The path planner has two components: sensor analytics and path finder. The sensor analytics component combines the uncertainties of all sensors to evaluate the positioning and navigation performance of an AV at given locations and times. The path finder component then utilizes the acquired sensor performance and creates a weight based on safety for each road segment. The operation and quality of the proposed path finder are demonstrated through simulations. The simulation results reveal that the proposed safe path planner generates paths that significantly improve the navigation safety in complex dynamic environments when compared to the paths generated by conventional approaches.

Download Full-text

Social Force Control for Human-Like Autonomous Driving

Volume 3: 20th International Conference on Advanced Vehicle Technologies; 15th International Conference on Design Education ◽

10.1115/detc2018-86288 ◽

2018 ◽

Author(s):

DoHyun Daniel Yoon ◽

Beshah Ayalew

Keyword(s):

Autonomous Vehicles ◽

Autonomous Vehicle ◽

Autonomous Driving ◽

Social Force ◽

Cruise Control ◽

Lane Changing ◽

Safe Navigation ◽

Proposed Model ◽

The Social ◽

Efficient Integration

An autonomous driving control system that incorporates notions from human-like social driving could facilitate an efficient integration of hybrid traffic where fully autonomous vehicles (AVs) and human operated vehicles (HOVs) are expected to coexist. This paper aims to develop such an autonomous vehicle control model using the social-force concepts, which was originally formulated for modeling the motion of pedestrians in crowds. In this paper, the social force concept is adapted to vehicular traffic where constituent navigation forces are defined as a target force, object forces, and lane forces. Then, nonlinear model predictive control (NMPC) scheme is formulated to mimic the predictive planning behavior of social human drivers where they are considered to optimize the total social force they perceive. The performance of the proposed social force-based autonomous driving control scheme is demonstrated via simulations of an ego-vehicle in multi-lane road scenarios. From adaptive cruise control (ACC) to smooth lane-changing behaviors, the proposed model provided a flexible yet efficient driving control enabling a safe navigation in various situations while maintaining reasonable vehicle dynamics.

Download Full-text

Safe navigation for autonomous vehicles: a purposive and direct solution

10.1117/12.150227 ◽

1993 ◽

Cited By ~ 1

Author(s):

Gin-Shu Young ◽

Tsai Hong Hong ◽

Martin Herman ◽

Jackson C. S. Yang

Keyword(s):

Autonomous Vehicles ◽

Direct Solution ◽

Safe Navigation

Download Full-text

Knowledge Acquisition from Pedestrian Flow Analysis using Sparse Mobile Probe Data

Journal of Intelligent & Robotic Systems ◽

10.1007/s10846-021-01419-w ◽

2021 ◽

Vol 102 (4) ◽

Author(s):

Ranulfo Plutarco Bezerra Neto ◽

Kazunori Ohno ◽

Thomas Westfechtel ◽

Shotaro Kojima ◽

Kento Yamada ◽

...

Keyword(s):

Autonomous Vehicles ◽

Flow Analysis ◽

Noisy Data ◽

Sensor Data ◽

Pedestrian Flow ◽

Trajectory Data ◽

Safe Navigation ◽

Daily Data ◽

New Information ◽

High Level

AbstractAutonomous vehicles require high-level semantic maps, which contain the activities of pedestrians and cars, to ensure safe navigation. High-level semantics can be obtained from mobile probe sensor data. Analyzing pedestrian trajectories obtained from mobile probe data is an effective approach to avoid collisions between autonomous vehicles and pedestrians. Such analyses of pedestrian trajectories can generate new information such as pedestrian behaviors in violation of traffic regulations. However, pedestrian trajectories obtained from mobile probe data significantly sparse and noisy, making it challenging to analyze pedestrian activity. To address this issue, we propose multiple daily data and graph-based approaches to treat sparse and noisy data for estimating the flow of pedestrians based on mobile probe data. To improve the sparseness of the data, multiple daily data are fused. After that, a pedestrian graph is created to enhance the region’s coverage by connecting the sparse data indicating the flow of pedestrians. This proposed approach successfully obtained pedestrian trajectory data from the sparse and noisy data. Moreover, it was possible to identify the potential locations where pedestrians tend to cross the street by analyzing the pedestrian flow. The results indicate that 83% of well-known regions where pedestrians tend to cross the street corresponded with those extracted using the proposed approach. Furthermore, a high-level semantic map of the regions where pedestrians tend to cross the street along a 1-km road is presented. The trajectory information obtained using the proposed approach is expected to be essential for understanding different scenarios of the interactions between individuals and autonomous vehicles.

Download Full-text