超声波传感器 机器人导航数据集

Data Set Information: 提供的文件包括三个不同的数据集。第一个包含所有24个超声波传感器的原始测量值和相应的类别标签。传感器读数的采样频率为每秒9个样本的速率。 第二个包含四个名为“简化距离”的传感器读数和相应的类别标签。这些简化的距离称为“前距离”、“左距离”、“右距离”和“后距离”。它们分别由位于机器人前部、左侧、右侧和后部60度弧内的传感器读数中的最小值组成。 第三个仅包含前面和左边的简化距离以及相应的类标签。 值得一提的是，24个超声波读数和简化的距离是在同一时间步采集的，因此每个文件具有相同的行数（每个采样时间步一行）。 “墙跟随”任务和数据收集旨在验证这样一个假设，即这个看似简单的导航任务确实是一个非线性可分离的分类任务。因此，线性分类器，如感知器网络，无法在不发生碰撞的情况下学习任务并指挥机器人在房间内走动。非线性神经分类器，如MLP网络，能够在不发生碰撞的情况下成功地学习任务和指挥机器人。 如果为神经分类器提供某种短期记忆机制，它们的性能通常会得到改善。例如，如果过去的输入与当前的传感器读数一起提供，则即使感知器也能够学习任务并成功地指挥机器人。如果使用递归神经网络（如Elman网络）来学习任务，则生成的动态分类器能够使用比MLP网络更少的隐藏神经元来学习任务。 建立了具有不同传感器读数数量的文件，以评估分类器在输入数量方面的性能。 Attribute Information: Number of Attributes -- sensor_readings_24.data: 24 numeric attributes and the class. -- sensor_readings_4.data: 4 numeric attributes and the class. -- sensor_readings_2.data: 2 numeric attributes and the class. For Each Attribute: -- File sensor_readings_24.data: 1. US1: ultrasound sensor at the front of the robot (reference angle: 180?°) - (numeric: real) 2. US2: ultrasound reading (reference angle: -165?°) - (numeric: real) 3. US3: ultrasound reading (reference angle: -150?°) - (numeric: real) 4. US4: ultrasound reading (reference angle: -135?°) - (numeric: real) 5. US5: ultrasound reading (reference angle: -120?°) - (numeric: real) 6. US6: ultrasound reading (reference angle: -105?°) - (numeric: real) 7. US7: ultrasound reading (reference angle: -90?°) - (numeric: real) 8. US8: ultrasound reading (reference angle: -75?°) - (numeric: real) 9. US9: ultrasound reading (reference angle: -60?°) - (numeric: real) 10. US10: ultrasound reading (reference angle: -45?°) - (numeric: real) 11. US11: ultrasound reading (reference angle: -30?°) - (numeric: real) 12. US12: ultrasound reading (reference angle: -15?°) - (numeric: real) 13. US13: reading of ultrasound sensor situated at the back of the robot (reference angle: 0?°) - (numeric: real) 14. US14: ultrasound reading (reference angle: 15?°) - (numeric: real) 15. US15: ultrasound reading (reference angle: 30?°) - (numeric: real) 16. US16: ultrasound reading (reference angle: 45?°) - (numeric: real) 17. US17: ultrasound reading (reference angle: 60?°) - (numeric: real) 18. US18: ultrasound reading (reference angle: 75?°) - (numeric: real) 19. US19: ultrasound reading (reference angle: 90?°) - (numeric: real) 20. US20: ultrasound reading (reference angle: 105?°) - (numeric: real) 21. US21: ultrasound reading (reference angle: 120?°) - (numeric: real) 22. US22: ultrasound reading (reference angle: 135?°) - (numeric: real) 23. US23: ultrasound reading (reference angle: 150?°) - (numeric: real) 24. US24: ultrasound reading (reference angle: 165?°) - (numeric: real) 25. Class: -- Move-Forward -- Slight-Right-Turn -- Sharp-Right-Turn -- Slight-Left-Turn -- File sensor_readings_4.data: 1. SD_front: minimum sensor reading within a 60 degree arc located at the front of the robot - (numeric: real) 2. SD_left: minimum sensor reading within a 60 degree arc located at the left of the robot - (numeric: real) 3. SD_right: minimum sensor reading within a 60 degree arc located at the right of the robot - (numeric: real) 4. SD_back: minimum sensor reading within a 60 degree arc located at the back of the robot - (numeric: real) 5. Class: -- Move-Forward -- Slight-Right-Turn -- Sharp-Right-Turn -- Slight-Left-Turn -- File sensor_readings_2.data: 1. SD_front: minimum sensor reading within a 60 degree arc located at the front of the robot - (numeric: real) 2. SD_left: minimum sensor reading within a 60 degree arc located at the left of the robot - (numeric: real) 3. Class: -- Move-Forward -- Slight-Right-Turn -- Sharp-Right-Turn -- Slight-Left-Turn Relevant Papers: Ananda L. Freire, Guilherme A. Barreto, Marcus Veloso and Antonio T. Varela (2009), 'Short-Term Memory Mechanisms in Neural Network Learning of Robot Navigation Tasks: A Case Study'. Proceedings of the 6th Latin American Robotics Symposium (LARS'2009), Valpara?-so-Chile, pages 1-6, DOI: 10.1109/LARS.2009.5418323 Citation Request: Please refer to the Machine Learning Repository's citation policy