Denavit-Hartenberg Parameters
- Introduced by
- Denavit, J. and Hartenberg, R. S.
- Year
- 1955
- Type
- Mathematical convention for describing articulated mechanisms
- Family
- Kinematic representation
What does it do?
The Denavit-Hartenberg (DH) convention is a systematic mathematical method for assigning coordinate frames to the links of an articulated robot or mechanism, enabling compact representation and computation of forward and inverse kinematics.
Introduced by Denavit and Hartenberg in 1955, this method uses only four parameters per joint to describe the spatial relationship between adjacent links, dramatically simplifying kinematic analysis and control of complex multi-jointed systems.
The intuition behind the method
The core insight is that any articulated mechanism can be described by assigning one coordinate frame per link and relating adjacent frames through a consistent set of four parameters: link length, twist angle, link offset, and joint angle. Rather than dealing with arbitrary 3D transformations, the DH convention imposes a rigid structure on the coordinate assignment, reducing the degrees of freedom and making the system manageable. Each joint contributes one variable parameter (the joint angle for revolute joints, or extension for prismatic joints), while the other three parameters are fixed geometry constants.
History
Denavit and Hartenberg introduced their convention in 1955 to systematize the analysis of linkage mechanisms. The method was adopted in the 1970s-1980s as industrial robotics emerged, and it has become the de facto standard in robotics textbooks and practice. Extensions such as the modified DH convention address limitations in certain configurations, but the original DH method remains most widely taught and applied.
When to use it
Use Denavit-Hartenberg parameters for any robotic or articulated system with multiple joints: industrial robots, robotic arms, gantry systems, or legged machines. It is essential for kinematics simulation, control system design, and programming. Assume the mechanism is rigid with well-defined joints; use this convention when designing or analyzing any articulated linkage.
Limitations and considerations
- Requires careful, consistent frame assignment; errors in parameter definition lead to incorrect kinematics
- Does not directly encode dynamic properties (mass, inertia) needed for control
- Parallel or non-standard joint configurations can be difficult to represent with DH convention
- Numerical singularities and gimbal lock can occur in certain configurations
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Compact, systematic representation requiring only four parameters per joint
- Enables efficient computation of forward kinematics using matrix multiplication
- Provides a standardized notation facilitating communication between researchers and engineers
- Simplifies inverse kinematics derivation for many common robot architectures
- Scales naturally to mechanisms with many degrees of freedom
Applications
- Industrial robotic arms: programming SCARA, articulated, and DELTA robots for manufacturing tasks
- Mobile robots: describing the kinematics of multi-legged walkers and wheeled platforms
- Surgical robotics: precise modeling of laparoscopic surgical instruments and da Vinci-type systems
- Humanoid robotics: analyzing the complex kinematics of bipedal and quadrupedal locomotion
How it works
How it works
Obtain the physical dimensions and configuration of the articulated mechanism: link lengths, joint axes, offsets between axes, and whether each joint is revolute (rotational) or prismatic (sliding).
Establish the base frame and the tool frame (end-effector); number the joints sequentially from base to tool.
For each joint, identify the common perpendicular between the joint axes of adjacent links and establish the orientation relationships.
Assign DH parameters for each joint: a_i (link length), alpha_i (twist angle), d_i (link offset), theta_i (joint angle). Build the homogeneous transformation matrix for each joint, and compose them to obtain the overall transformation from base to tool.
Use the transformation matrices to compute forward kinematics (tool position and orientation given joint angles) or solve inverse kinematics (desired joint angles for a target tool pose).
Common pitfalls
- Inconsistent frame assignment conventions (modified DH vs. standard DH) leading to parameter incompatibility
- Confusing joint angles with link parameters, resulting in incorrect kinematic models
- Neglecting to account for the order of operations (rotations before translations) in the DH transformation sequence
Frequently asked questions
What is the difference between the standard DH convention and the modified DH convention?
The standard DH convention assigns the z-axis along the current joint axis, while the modified version assigns it along the next joint axis. Modified DH is sometimes preferred for clarity in representing certain mechanisms, but both are equivalent; choose one and be consistent.
How do I assign DH parameters for a planar 2-link arm?
For a 2-link planar arm with revolute joints: frame 0 at the base, frame 1 at joint 2, frame 2 at the end-effector. Parameters are: a1 (length of link 1), a2 (length of link 2), with zero offset and twist angles. Joint angles theta1 and theta2 are the variables.
What happens if two consecutive joint axes are parallel?
Parallel axes have zero twist angle (alpha = 0). The link offset d then determines the perpendicular distance between the axes. This is a common configuration and the DH convention handles it naturally.
Can I use DH parameters for a robot with closed-loop kinematic chains?
DH parameters are designed for open-chain mechanisms (trees). Closed loops require additional constraints and are typically analyzed using constraint equations or by treating them as redundant open chains with constraints.
Related methods
Sources
Denavit, J., & Hartenberg, R. S. (1955). A kinematic notation for lower-pair mechanisms based on matrices. Journal of Applied Mechanics, 22(2), 215-221.
Open publication referenceCopied.Craig, J. J. (2005). Introduction to Robotics: Mechanics and Control (3rd ed.). Pearson Education.
ISBN: 0-13-123629-6Copied.Spong, M. W., Hutchinson, S., & Vidyasagar, M. (2006). Robot Modeling and Control. John Wiley & Sons.
ISBN: 0-471-64990-2Copied.