포인트 클라우드를 이용한 충돌회피 알고리즘
Collision Avoidance Algorithm Using Point Cloud
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
본 논문에서는 유니티 가상공간에서 포인트 클라우드를 활용한 충돌회피 알고리즘을 제안한다. 충돌회피를 위한알고리즘은 먼저, 카메라의 전방을 기준으로 랜덤하게 레이를 발사하여 포인트 클라우드를 획득하고, 이를 바탕으로바닥과 그 이외의 영역을 추출한다. 추출된 바닥에서 설정된 목표점까지의 방향 벡터를 생성하고, 이 벡터를 기준으로가장 먼 포인트까지 갈 수 있는 경로를 A* 알고리즘을 사용하여 구성한다. 이후, 경로상의 각 웨이포인트(waypoint)마다 주변 장애물들과의 거리를 계산하여 일정 거리 이하의 웨이포인트를 제거하는 필터링 과정을 거친다. 필터링된 웨이포인트를 따라 에이전트는 이동하며, 이동 중 좌우로 설정된 각도에서 레이를 방출하여 충돌 가능성이 있는 물체를 감지한다. 충돌 물체가 감지되면, 감지된 방향과 반대 방향으로 회전하여 웨이포인트를 따라 계속 이동함으로써 장애물을회피하며 목표 지점까지 도달할 수 있다. 이 알고리즘은 실시간 환경에서 장애물 회피와 경로 최적화를 결합한 효율적인 경로 계획 방법을 제시하며, 자율주행 및 로봇 내비게이션 시스템에 적용 가능할 것이다.
Abstract
This paper proposes a collision avoidance algorithm using point clouds in a virtual environment created in Unity. First, random rays are fired from the front of a camera to acquire a point cloud, which is then used to extract the floor and other regions. A direction vector towards the target point is generated from the extracted floor, and based on this vector, a path to the farthest point is constructed using the A* algorithm. Next, each waypoint along the path is filtered by calculating the distance to nearby obstacles, removing waypoints that are too close. The agent follows the filtered waypoints while emitting rays at set angles to the left and right to detect potential obstacles. If an obstacle is detected, the agent rotates in the opposite direction and continues along the waypoint path to avoid the obstacle and reach the target. This algorithm presents an efficient method for combining obstacle avoidance and path optimization in real-time environments, with strong applicability in autonomous driving and robotic navigation systems.
참고문헌 (21)
- Y. Peng, "The obstacle detection and obstacle avoidance algorithm based on 2-d lidar", Proc. 2015 IEEE Int. Conf. Information and Automation, pp.1648~1653 (2015)
- A. Shitsukane, "A survey on obstacles avoidance mobile robot in static unknown environment", Int. J. Comput, 28(1), pp.160~173 (2018)
- Y. Cao, "An improved dynamic window approach algorithm for dynamic obstacle avoidance in mobile robot formation", Decision Analytics Journal, 11, pp.100471~ (2024)
- E. J. Molinos, "Dynamic window based approaches for avoiding obstacles in moving", Robotics and Autonomous Systems, 118, pp.112~130 (2019)
- K. Shibata, "Model predictive obstacle avoidance control for vehicles with automatic velocity suppression using artificial potential field", IFAC-PapersOnLine, 51(20), pp.313~318 (2018)
- O. Khatib, "Real-time obstacle avoidance for manipulators and mobile robots", The International Journal of Robotics Research, 5(1), pp.90~98 (1986)
- C. R. Qi, "Pointnet : Deep learning on point sets for 3d classification and segmentation", Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp.652~660 (2017)
- C. R. Qi, "Pointnet++ : Deep hierarchical feature learning on point sets in a metric space", Advances in Neural Information Processing Systems, 30 (2017)
- Y. Wang, "Dynamic graph CNN for learning on point clouds", ACM Trans. Graphics(TOG), 38(5), pp.1~12 (2019)
- T. Gomes, "A survey on ground segmentation methods for automotive LiDAR sensors", Sensors, 23(2), pp.601~ (2023)
- B. Wang, "LiDAR filtering in 3D object detection based on improved RANSAC", Remote Sensing, 14(9), pp.2110~ (2022)
- C. Chen, "Segmentation-based hierarchical interpolation filter using both geometric and radiometric features for LiDAR point clouds over complex scenarios", Measurement, 211, pp.112668~ (2023)
- L. Chu, "Intelligent vehicle path planning based on optimized A* algorithm", Sensors, 24(10), pp.3149~ (2024)
- R. Yonetani, "Path planning using neural A*search", Proc. Int. Conf. Machine Learning, pp.12029~12039 (2021), PMLR
- M. Yermo, "A fast and optimal pathfinder using airborne LiDAR data", ISPRS Journal of Photogrammetry and Remote Sensing, 183, pp.482~495 (2022)
- K. Al-Mutib, "D* lite based real-time multi-agent path planning in dynamic environments", 2011 Third International Conference on Computational Intelligence, Modelling & Simulation, pp.170~174 (2011), IEEE
- M. Kamezaki, "Dynamic waypoint navigation : model-based adaptive trajectory planner for human-symbiotic mobile robots", IEEE Access, 10, pp.81546~81555 (2022)
- S. J. Al-Kamil, "Enhancing mobile robot navigation : Optimization of trajectories through machine learning techniques for improved path planning efficiency", Mathematics, 12(12), pp.1787~ (2024)
- C. Ericson, "Real-time Collision Detection" (2004), CRC Press
- D. Meister, "A survey on bounding volume hierarchies for ray tracing", Computer Graphics Forum, 40(2), pp.683~712 (2021)
- H. A. Sulaiman, "Bounding volume hierarchies for collision detection", Computer Graphics, 10, pp.5772~2386 (2012)
