Dark Mode
Industrial Robot Kinematics Dataset
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset offers kinematic data for an ABB IRB 120 robot arm, providing a clear link between its joint angles and the resultant end-effector positions. Robotic manipulator inverse kinematics, which is the process of determining the required joint positions to achieve a specific end-effector location, can be analytically challenging. In contrast, direct kinematics, which maps joint-coordinate space to tool-coordinate space (x, y, z coordinates relative to the robot's base), is simpler to ascertain. This dataset, generated using direct kinematic equations, provides 15,000 such mappings, making it an ideal resource for machine learning research aimed at predicting inverse kinematics, potentially leading to more efficient or accurate solutions than traditional analytical methods. Researchers have already used this type of dataset for regression with Multilayer Perceptrons and Genetic Programming algorithms.
Columns
- q1: Represents the angle of the first joint of the robotic manipulator, generated uniformly at random within its possible range. Its values span from -2.88 to 2.88 radians, with a mean close to 0.
- q2: Represents the angle of the second joint of the robotic manipulator, generated uniformly at random. Its values span from -1.92 to 1.92 radians, with a mean close to -0.01.
- q3: Represents the angle of the third joint of the robotic manipulator, generated uniformly at random. Its values span from 1.22 to 1.92 radians, with a mean close to 1.57.
- q4: Represents the angle of the fourth joint of the robotic manipulator, generated uniformly at random. Its values span from -2.79 to 2.79 radians, with a mean close to 0.02.
- q5: Represents the angle of the fifth joint of the robotic manipulator, generated uniformly at random. Its values span from -2.09 to 2.09 radians, with a mean close to -0.01.
- q6: Represents the angle of the sixth joint of the robotic manipulator, generated uniformly at random. Its values span from 0.00 to 6.28 radians, with a mean close to 3.14.
- x: The X-position of the robot's end-effector, calculated using the direct kinematics model based on the corresponding joint angles. Its values span from -0.29 to 0.29, with a mean close to -0.01.
- y: The Y-position of the robot's end-effector, calculated through the direct kinematics model. Its values span from -0.30 to 0.30, with a mean close to 0.
- z: The Z-position of the robot's end-effector, calculated through the direct kinematics model. Its values span from -0.02 to 0.49, with a mean close to 0.26.
Distribution
The dataset is provided as a data file, typically in CSV format, named
robot_inverse_kinematics_dataset.csv. It consists of 15,000 datapoints. Each datapoint comprises a pair of vectors: six joint coordinates (angles in radians) and the corresponding x, y, z coordinates of the tool-coordinate space. The file size is 1.28 MB.Usage
This dataset is particularly suitable for regression tasks and machine learning research focused on determining the inverse kinematics of robotic manipulators. Specific applications include:
- Developing and testing new regression methods for predicting robot joint positions from end-effector coordinates.
- Comparing the efficacy of various machine learning algorithms, such as Multilayer Perceptrons and Genetic Programming, for inverse kinematics problems.
- Exploring data-driven approaches to overcome the analytical challenges often associated with inverse kinematics.
- Educational use in robotics, kinematics, and machine learning courses.
Coverage
This dataset represents a simulated environment for the ABB IRB 120 robot arm, an industrial robotic manipulator. The data pertains to the kinematic behaviour of this specific model. The data is synthetic, generated by uniformly randomising joint values within their permitted ranges. Consequently, there is no specific geographic, time range, or demographic scope associated with this dataset.
License
Attribution 4.0 International (CC BY 4.0)
Who Can Use It
This dataset will be useful for a diverse group of users, including:
- Robotics Engineers and Researchers: For developing and evaluating algorithms for robot control, path planning, and inverse kinematics solutions.
- Machine Learning Practitioners: To apply and test different regression models on an engineering challenge.
- Students and Academics: As a practical case study for academic pursuits in robotics, control systems, and artificial intelligence.
- Data Scientists: To explore data generation techniques and model complex relationships within physical systems.
Dataset Name Suggestions
- IRB 120 Robot Kinematics Data
- Robotic Arm Inverse Kinematics Dataset
- ABB IRB 120 Joint to Tool Space Data
- Robot Kinematics Regression Data
- Industrial Robot Kinematics Dataset
Attributes
Original Data Source: Industrial Robot Kinematics Dataset
QUALITY