The global demand for dairy products has driven significant advancements in livestock farming technology, with automated milking systems (AMS) emerging as a pivotal innovation for enhancing efficiency and product quality. The performance of the robotic manipulator, particularly its end effector, is a critical determinant of the system’s overall stability and milking throughput. This component is responsible for the precise tasks of locating, cleaning, and attaching teat cups to the cow’s udder. The design and functionality of the milking robot’s end effector directly impact the speed of operation and its ability to adapt to the natural variations in udder morphology and teat positioning among dairy cattle.
Existing commercial milking robot end effectors primarily follow two design philosophies. The first employs a single-gripper end effector mounted on a multi-axis robotic arm. While offering high dexterity, this design requires sequential attachment of four teat cups, leading to prolonged milking cycles and reduced overall efficiency. The second type integrates all four teat cups directly into a simpler, often 3-DOF, end effector structure, enabling simultaneous attachment. However, this integrated design typically lacks individual adjustment mechanisms for each cup, resulting in poor adaptability to varying teat distances and angles, which in turn causes attachment failures and inefficiency.
This study addresses these shortcomings by applying the Theory of Inventive Problem Solving (TRIZ) to systematically innovate and design a novel, adaptive, dual-gripper end effector for milking robots. The goal is to develop an end effector that balances high operational speed with excellent adaptability, ultimately improving milking robot performance.
Problem Analysis and Root Cause Identification
The initial step involves a thorough problem definition. The primary issues observed in current milking robot operations are low attachment efficiency and poor adaptability of the end effector during the teat cup positioning (cupping) process. To move beyond symptomatic understanding, a Cause-Effect Chain Analysis, a core TRIZ tool, was conducted. This analysis traces the chain of logical connections from observed undesirable effects back to their fundamental root causes.
The analysis identified a core chain leading to low efficiency: The immediate cause was identified as an excessive number of cup pickup and attachment cycles required per cow. Tracing this back revealed that this was a direct result of having an insufficient number of grippers on the end effector. A parallel chain leading to poor adaptability was also mapped: The immediate cause was the end effector’s inability to compensate for varied teat positions. The root causes for this were found to be the gripper’s lack of degrees of freedom (DOF), specifically the absence of mechanisms for longitudinal and lateral angular adjustments. Furthermore, the frequent need to change grippers or cups between operations was also flagged as a contributing factor to time loss. This causal analysis conclusively pinpointed the fundamental design flaws: an inadequate number of grippers and a lack of micro-adjustment capability in the existing end effector designs.
TRIZ-Based Innovative Design of the End Effector
Resolving Technical Contradictions for Structural Innovation
The identified root causes were translated into standard TRIZ engineering parameters to apply the Contradiction Matrix. Two key contradictions were formulated:
- Improving “Adaptability and Versatility” of the end effector leads to the worsening of “Device Complexity.”
- Improving “Loss of Time” (increasing speed) leads to the worsening of “Stability of the Object’s Structure.”
Consulting the Altshuller’s Contradiction Matrix provided the following inventive principles for each contradiction pair, as summarized in the table below:
| Improving Parameter | Worsening Parameter | Suggested Inventive Principles |
|---|---|---|
| Adaptability (35) | Device Complexity (36) | 15 (Dynamicity), 29 (Pneumatics/Hydraulics), 37 (Thermal Expansion), 28 (Mechanics Substitution) |
| Loss of Time (25) | Stability (13) | 35 (Parameter Changes), 3 (Local Quality), 22 (Blessing in Disguise), 5 (Merging) |
| Loss of Time (25) | Device Complexity (36) | 6 (Universality), 29 (Pneumatics/Hydraulics) |
From this set, principles #5 (Merging) and #15 (Dynamicity) were selected as the most directly applicable for the end effector redesign.
- Principle #5 (Merging): Application 1: “Merge identical or related objects in space.” This principle directly suggests combining multiple single grippers into one unit. Consequently, a dual-gripper configuration was adopted for the new end effector. This allows the robot to pick up and position two teat cups simultaneously, effectively halving the number of major motion cycles required compared to a single-gripper design.
- Principle #15 (Dynamicity): Application 3: “Make a rigid or fixed object movable or adaptive.” To address the adaptability issue, micro-adjustment mechanisms were introduced at the base of each gripper on the end effector. This includes two independent angular adjustment actuators (e.g., servo motors) per gripper to provide precise pitch (longitudinal) and yaw (lateral) orientation changes. Furthermore, a lead-screw guide rail mechanism with left-handed and right-handed threads was integrated between the two grippers. Actuation of this screw simultaneously adjusts the distance between the two grippers, allowing the end effector to adapt to varying teat spacings. This dynamic adjustability occurs during the final, fine-positioning phase after the robot arm has placed the end effector in the general udder area.
The key parameters for the designed gripper unit are as follows:
| Parameter | Value / Specification |
|---|---|
| Payload | ≥ 20 N |
| Mass | ≤ 1.4 kg |
| Dimensions (L×W×H) | 113 × 67 × 40 mm |
| Axial Positioning Error | ≤ 1 mm |
| Gripping Force | ≥ 34 N |
| Teat Cup Compatibility | Radius 15 mm, Height 120 mm |
Substance-Field Analysis for Actuation System Optimization
While the structural concept was defined, the choice of actuation for the gripper itself needed optimization. The initial design considered an electric gripper. A Substance-Field (Su-Field) analysis was performed to model this system. In the Su-Field model, the minimum technical system consists of two substances (S1, S2) and a field (F) that mediates the interaction between them.
- S1: The Teat Cup (Object to be manipulated).
- S2: The Electric Gripper (Tool).
- F: Electro-Mechanical Field (provides the gripping force and control).
The model revealed an “Insufficient Effect”: In the damp, wash-down environments typical of dairy operations, prolonged exposure can degrade the precision and reliability of electric components. This could lead to inconsistent gripping force or positional drift in the electric end effector gripper, causing cup slippage or misalignment, which reduces efficiency and could harm the animal.
To resolve this insufficient effect, the standard solution of “Replacing the Field” was applied. The electro-mechanical field (F1) was replaced with a pneumatic field (F2). Consequently, S2 was changed from an electric gripper to a pneumatic gripper.
Advantages of the Pneumatic Solution for the End Effector:
- Environmental Robustness: Pneumatic systems are inherently more resistant to moisture and wash-down chemicals, increasing reliability and lifespan.
- Simplicity & Speed: Pneumatic grippers have a simpler mechanical structure and can operate at higher speeds, reducing the time for cup grasping and release cycles.
- Inherent Compliance & Safety: The compressibility of air provides a degree of natural compliance, which can be beneficial for delicate handling tasks and offers inherent safety.
- Reduced On-Board Mass: The actuator (air cylinder) can be made very compact and lightweight. The primary power source (compressor) can be located remotely, reducing the mass on the moving end effector and improving the robot arm’s dynamic performance.
The final conceptual design of the innovative end effector integrates a dual-gripper configuration, with each gripper being pneumatically actuated and mounted on a platform featuring two angular adjustment servos. The two gripper platforms are connected via a central lead-screw mechanism for width adjustment, all mounted on the robot arm’s flange.
Kinematic Analysis and Workspace Verification
To validate the functional capability of the proposed dual-gripper end effector, a kinematic analysis was performed. Since the two grippers are symmetric, the analysis focuses on one side of the end effector. The kinematic chain for one gripper consists of one prismatic joint (the width adjustment rail) and two revolute joints (the pitch and yaw adjustment axes), forming a PRR serial manipulator from the base of the adjustment mechanism to the gripper’s contact point.
Denavit-Hartenberg (D-H) Modeling
The D-H convention is used to systematically describe the geometry and kinematics of the end effector’s adjustment mechanism. Frames are assigned to each joint, and parameters are defined as per the standard D-H method. The homogeneous transformation matrix between consecutive frames i-1 and i is given by:
$$
^{i-1}T_i = \begin{bmatrix}
\cos\theta_i & -\sin\theta_i\cos\alpha_{i-1} & \sin\theta_i\sin\alpha_{i-1} & a_{i-1}\cos\theta_i \\
\sin\theta_i & \cos\theta_i\cos\alpha_{i-1} & -\cos\theta_i\sin\alpha_{i-1} & a_{i-1}\sin\theta_i \\
0 & \sin\alpha_{i-1} & \cos\alpha_{i-1} & d_i \\
0 & 0 & 0 & 1
\end{bmatrix}
$$
The D-H parameters for one side of the end effector’s adjustment mechanism are established as follows:
| Link i | $a_{i-1}$ (mm) | $\alpha_{i-1}$ (deg) | $d_i$ (mm) | $\theta_i$ (deg) | Variable Range |
|---|---|---|---|---|---|
| 1 | 0 | 0 | 10 | $\theta_1$ | -90 to +90 |
| 2 | 99 | -90 | $d_2$ | -90 | 29 to 59 mm |
| 3 | 68 | -90 | 0 | $\theta_3$ | -50 to 0 |
| 4 | 0 | -90 | 47 | $\theta_4$ | -60 to +60 |
| 5 | -117 | 0 | 52 | 0 | — |
The overall transformation from the base frame (0) to the gripper’s tool point frame (5) is obtained by successive multiplication of the individual transformation matrices:
$$
^{0}T_5 = ^{0}T_1 \cdot ^{1}T_2 \cdot ^{2}T_3 \cdot ^{3}T_4 \cdot ^{4}T_5
$$
The resulting matrix $^{0}T_5$ has the form:
$$
^{0}T_5 = \begin{bmatrix}
n_x & o_x & a_x & p_x \\
n_y & o_y & a_y & p_y \\
n_z & o_z & a_z & p_z \\
0 & 0 & 0 & 1
\end{bmatrix} = [\mathbf{n} \quad \mathbf{o} \quad \mathbf{a} \quad \mathbf{p}]
$$
Where the vector $\mathbf{p} = [p_x, p_y, p_z]^T$ represents the position of the gripper’s tool point in the base coordinate system, which is the primary output for workspace analysis.
Workspace Simulation and Validation
The reachable workspace of the gripper point was determined using a Monte Carlo method. Random values for each joint variable ($\theta_1$, $d_2$, $\theta_3$, $\theta_4$) were generated within their specified limits using the following equations, where $N$ is the number of random samples (e.g., 10,000), and $rand(N,1)$ generates uniformly distributed random numbers between 0 and 1.
$$
\theta_i = \theta_{i}^{min} + (\theta_{i}^{max} – \theta_{i}^{min}) \cdot rand(N,1)
$$
$$
d_i = d_{i}^{min} + (d_{i}^{max} – d_{i}^{min}) \cdot rand(N,1)
$$
These random joint values were substituted into the forward kinematics equation to compute the corresponding $(p_x, p_y, p_z)$ coordinates for the end effector tool point. The aggregation of these points defines the mechanism’s reachable workspace. The simulation was performed using numerical computing software.
The resulting 3D and 2D projection plots of the workspace reveal a comprehensive operating volume. Analysis shows that the end effector’s gripper point can reach within a space defined approximately by: $-55 \text{ mm} < p_x < 160 \text{ mm}$, $-30 \text{ mm} < p_y < 230 \text{ mm}$, and $-100 \text{ mm} < p_z < 250 \text{ mm}$, relative to its base on the adjustment mechanism.
This workspace is then evaluated against the biological constraints of dairy cattle. Typical teat geometric parameters for common breeds are:
| Parameter | Typical Range |
|---|---|
| Teat Spacing (between adjacent teats) | 80 – 120 mm |
| Teat Length | 65 – 85 mm |
| Teat Diameter | 20 – 25 mm |
The analysis confirms that the designed end effector’s adjustment mechanism, with its combined rail and angular joints, provides a sufficient range of motion to cover the expected variations in teat positions for successful cup attachment. The workspace envelops the required operational volume, validating the kinematic design of the adaptive end effector.
Conclusion
This study demonstrates the effective application of TRIZ theory as a structured methodology for the innovative design of a robotic end effector in agricultural automation. By systematically applying tools like Cause-Effect Chain Analysis, the Technical Contradiction Matrix, and Substance-Field Analysis, fundamental problems of low efficiency and poor adaptability in existing milking robot end effectors were addressed.
The proposed solution is a novel, pneumatically-actuated, dual-gripper end effector with integrated micro-adjustment capabilities. The design merges two grippers to reduce operational cycles and incorporates dynamic adjustment mechanisms—a lead-screw rail for inter-gripper spacing and dual-axis angular servos per gripper—to adapt to variable teat geometries. The substitution of electric with pneumatic actuation enhances reliability in harsh farm environments.
Kinematic modeling and workspace simulation verify that the designed end effector possesses a sufficient range of motion to perform the required milking tasks. The TRIZ-driven approach not only solved the specific technical contradictions but also led to a holistic and optimized end effector design. This innovative end effector has the potential to significantly improve the speed, reliability, and adaptability of automated milking systems, contributing to more efficient and sustainable dairy farming operations. The methodological framework also provides a valuable reference for the development of other robotic manipulators and end effectors facing similar multi-objective design challenges.