In recent years, the presence of autonomous robots in the real world, for example, self-driving cars, drones, and unmanned surface vehicles have significantly increased. With the recent advances in machine/deep learning, there are growing expectations that full autonomy may become a reality shortly, and it is expected to bring fundamental changes to the societies of robotics, computer vision, and artificial intelligence.

An autonomous system typically consists of a series of modules comprising perception, navigation, planning, and control. The perception system is responsible for estimating location and constructing the 3-D environment map to plan safe navigation routes. With recent advances in machine/deep learning, such as convolutional neural networks, autonomous robots’ perception, navigation, and planning, robots have become more intelligent than ever before, and such systems' applications are being realized.

This Research Topic aims to present current directions in this field and explores the problems related to machine vision and intelligence for autonomous systems in the real world. Specifically, this Research Topic will mainly focus on:

1. Affordable sensors for varying environmental conditions;
2. Reliable simultaneous localization and mapping;
3. Machine learning that can effectively handle varying real-world conditions and unforeseen events;
4. Hardware and software co-design for efficient real-time performance;
5. Resilient and robust platforms that can withstand adversarial attacks and failures;
6. End-to-end system integration of sensing, computer vision, signal/image processing, and machine/deep learning.

In this way, relevant themes for this Research Topic include, but are not limited to:

• 3D environment reconstruction and understanding;
• Mapping and localization for unmanned vehicles in the real world;
• Semantic/instance segmentation and semantic mapping;
• Self-supervised/unsupervised visual environment perception;
• Obstacle detection/tracking and 3D localization;
• Signage detection and recognition;
• Deep/machine learning and image analysis for intelligent environment perception;
• Adversarial domain adaptation for autonomous systems;
• On-board embedded visual perception systems;
• Bio-inspired vision sensing for autonomous system perception;
• Real-time deep learning inference.

Keywords: 

Perception, navigation, planning, unmanned vehicles, AI

Important Note: 

All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.

Guest Editors:

Rui Fan, Tongji University

Nan Li, Northwestern Polytechnical University

Mohammud J. Bocus, University of Bristol

Yuxiang Sun, Hong Kong Polytechnic University

Yue Wang, Zhejiang University

Contact: 

Rui Fan, rui.fan@ieee.org