全向移動平台動態物件追蹤控制與設計
No Thumbnail Available
Date
2023
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
全向移動平台（Omnidirectional Mobile Platform）是具有全向性的移動機器人平台，其機動性更高，但相對於傳統四輪平台在運動控制上也更為複雜。本文自行設計此移動平台，從車身、驅動系統、控制系統、各種感測器以及影像識別系統，並且整合移動平台和影像系統的軟、硬體功能。使其能完成一些簡單擬人類的動作。接著，詳細說明了全向移動平台的運動學分析，包括其運動模型和運動方程式。在此基礎上，介紹了運動控制方法，包括階層模糊控制和由PID控制結合類神經網路的PIDNN控制，並且加上動態物件追蹤技術判斷目標位置，並且預測其落點。最後，實驗結果表明，所提出的控制方法可以實現對平台的運動控制和穩定性控制，並且可以有效追蹤物件並且移動到指定定點。此外，該方法也具有一定的通用性，可應用於其他需要追蹤物件的全向移動平台上。
Omnidirectional Mobile Platform is an omnidirectional mobile robot with higher mobility, but it’s also more complex in motion control than traditional four-wheel platforms. This thesis designs the mobile platform by myself, from drive system, control system, various sensors, and integrates the software and hardware functions of the mobile platform and image recognition system.Next, the kinematics analysis of the omnidirectional mobile platform is described. On this basis, the motion control method is introduced, including hierarchical fuzzy control and PIDNN control combined with PID control and neural network, and combined with dynamic object tracking technology to detect target position and predict its landing point.Finally, the experimental results show that the proposed control method can realize the motion control of the platform, and can effectively track the specific object and move to target. In addition, the method has certain versatility and can be applied to other omnidirectional mobile platforms that need to track object.
Omnidirectional Mobile Platform is an omnidirectional mobile robot with higher mobility, but it’s also more complex in motion control than traditional four-wheel platforms. This thesis designs the mobile platform by myself, from drive system, control system, various sensors, and integrates the software and hardware functions of the mobile platform and image recognition system.Next, the kinematics analysis of the omnidirectional mobile platform is described. On this basis, the motion control method is introduced, including hierarchical fuzzy control and PIDNN control combined with PID control and neural network, and combined with dynamic object tracking technology to detect target position and predict its landing point.Finally, the experimental results show that the proposed control method can realize the motion control of the platform, and can effectively track the specific object and move to target. In addition, the method has certain versatility and can be applied to other omnidirectional mobile platforms that need to track object.
Description
Keywords
全向移動平台, 階層模糊控制, PIDNN控制, 物件追蹤, Omnidirectional mobile, PIDNN control, Object tracking