[논문 리뷰] Mobile Robot Path Planning in Dynamic Environments: A Survey

Kuanqi Cai, Chaoqun Wang외 3명|arXiv (Cornell University)|2020. 06. 25.

Robotic Path Planning Algorithms참고 문헌 53인용 수 43

한 줄 요약

이 논문은 밀집된 동적 환경에서 이동 로봇의 경로 계획 방법을 조사하며, 글로벌 대 로컬 계획, Velocity Obstacle 방법들, 그리고 강화 학습 접근법을 강조한다.

ABSTRACT

There are many challenges for robot navigation in densely populated dynamic environments. This paper presents a survey of the path planning methods for robot navigation in dense environments. Particularly, the path planning in the navigation framework of mobile robots is composed of global path planning and local path planning, with regard to the planning scope and the executability. Within this framework, the recent progress of the path planning methods is presented in the paper, while examining their strengths and weaknesses. Notably, the recently developed Velocity Obstacle method and its variants that serve as the local planner are analyzed comprehensively. Moreover, as a model-free method that is widely used in current robot applications, the reinforcement learning-based path planning algorithms are detailed in this paper.

연구 동기 및 목표

• Identify challenges in navigation within densely populated dynamic environments.
• Frame mobile robot navigation into global and local planning to assess executability.
• Summarize recent progress and compare strengths/weaknesses of major methods.
• Highlight the Velocity Obstacle family and reinforcement learning as prominent approaches.

제안 방법

• Describe the navigation framework for mobile robots with global and local planning scopes.
• Survey recent path planning methods and categorize them by planning level and technique.
• Analyze the Velocity Obstacle method and its variants for local planning.
• Detail model-free reinforcement learning-based path planning approaches.
• Discuss strengths and weaknesses of surveyed methods in dense dynamic environments.

실험 결과

연구 질문

• RQ1 What are the recent progress and limitations of path planning methods for mobile robots in dense dynamic environments?
• RQ2 How do global and local planning approaches compare in terms of executability and performance?
• RQ3 What roles do Velocity Obstacle methods and their variants play in local planning?
• RQ4 How are reinforcement learning-based path planning methods applied and what are their strengths/weaknesses?
• RQ5 What are key challenges and future directions identified for dynamic-environment navigation?

주요 결과

• Velocity Obstacle methods and their variants are analyzed comprehensively as local planners.
• Reinforcement learning-based path planning methods are detailed as model-free approaches widely used in current applications.
• The survey discusses strengths and weaknesses of various planning methods within dense dynamic environments.