The pursuit of agile and efficient mobile robots for challenging terrains remains a significant focus in robotics research. Nature, through millennia of evolution, offers exemplary solutions for locomotion. Among these, the explosive jump of the frog, particularly species like the black-spotted frog (Pelophylax nigromaculatus), presents a compelling model. Its ability to propel itself over distances exceeding ten times its body length through a rapid release of stored elastic energy is a feat of biomechanical engineering. This study is motivated by the goal of replicating this powerful and efficient jumping gait in a mechanical system. The core contribution lies in the innovative conceptualization and detailed design of a novel bionic robot centered around a six-bar linkage energy storage mechanism and a unique rope-based locking-release system. This design philosophy prioritizes structural simplicity, high energy density, and reliable triggering, addressing common challenges in jump-capable bionic robot platforms such as complex actuation, poor energy conversion, and unstable take-off postures.
Jumping as a mobility mode offers distinct advantages for small-scale robots, enabling them to overcome obstacles significantly larger than their own size, traverse gaps, and navigate rough terrain quickly. Several research efforts have explored bionic robot designs inspired by frogs. These range from complex multi-degree-of-freedom leg mechanisms to simpler spring-loaded catapult systems. While some designs successfully mimic the kinematic sequence, they often suffer from mechanical complexity, high weight, or inefficient energy transfer from the actuator to the jump. Our approach simplifies the biomimetic principle into a robust mechanical embodiment. We focus on the core function—energy storage and rapid release—by designing a constrained six-bar linkage that transforms linear spring deflection into a coordinated leg extension, mimicking the frog’s preparatory crouch and powerful push-off. This paper details the comprehensive design process, from the mechanical architecture and static analysis to the kinematic and dynamic modeling of the proposed bionic robot. The integration of a topology-optimized structure and a minimalist wire-triggered release mechanism aims to achieve a high performance-to-complexity ratio, paving the way for practical applications in exploration and reconnaissance.
Mechanical Architecture of the Bionic Jumping Robot
The overall design of the bionic robot is a synthesis of several integrated subsystems. The primary objective was to create a compact, self-contained system capable of repeated, stable jumps. The key innovation resides in the leg mechanism, which departs from traditional rotational joint-based designs or simple prismatic springs.
System Overview and Subassemblies
The robot can be decomposed into five main functional groups: the chassis and forelegs, the drive unit, the six-bar energy storage mechanism, the hind feet, and the locking-release module. The chassis, constructed from a primary mounting plate, provides the foundational frame. The forelegs are passive, attached via hinged connections that allow them to rotate and adapt their angle during the compression and launch phases, contributing to a stable pre-jump posture and landing. The drive unit consists of a geared DC motor fixed to the chassis. The heart of the bionic robot is the six-bar linkage, which forms the reconfigurable structure of the hind limbs. Finally, the locking-release module, based on an incomplete gear and a wound cable, controls the critical transition from energy loading to energy release.
Innovative Six-Bar Energy Storage Mechanism
Most jumping mechanisms utilize either a direct linear spring or a rotational spring coupled with a lever. Our design employs a planar six-bar linkage to couple the motion of two tension springs to the leg extension. This mechanism, illustrated in the system schematic, consists of two thigh links, two shank links, and a pair of intermeshing synchronized gears. These components are connected in a closed-loop configuration via pin joints.
- Thigh and Shank Links: These are the primary structural members. They are connected to each other at the “knee” joint and to the chassis and foot attachment points at the “hip” and “ankle,” respectively.
- Synchronized Gears: A pair of gears, mounted on the chassis, interconnect the two symmetric sides of the linkage. This gear pair is crucial as it kinematically couples the left and right leg mechanisms, forcing them to move symmetrically. This coupling reduces the system’s degrees of freedom to one, ensuring a deterministic and repeatable motion path. The synchronized motion is vital for a stable, straight jump, preventing unwanted yaw or roll during take-off.
- Energy Storage Springs: Two tension springs are connected between a specific joint on the linkage (the “connector” in the schematic) and a fixed point on the chassis. As the linkage is deformed from an extended (“straight leg”) configuration to a compressed (“crouched”) configuration, these springs are stretched, storing elastic potential energy.
- Topology Optimization: To maximize strength while minimizing weight—a critical factor for jump height—the thigh and shank links were subjected to static stress analysis under expected launch loads. Using topology optimization algorithms, material was strategically removed from low-stress regions, resulting in an organic, bone-like truss structure for these components. This process significantly enhances the structural efficiency of the bionic robot.
The kinematics of this mechanism are described by the following geometric constraints, where \( L \) is the length of each thigh/shank link (assumed equal for symmetry), \( \theta \) is the acute angle each link makes with the horizontal axis in a symmetric configuration, \( W_c \) is the center distance between the gear axes (fixed chassis width), and \( H_c \) is the compressed height of the mechanism.
The horizontal span during compression is given by the sum of the horizontal projections of the four links plus the chassis width:
$$ W_{span} = 2L \cos\theta + 2L \cos\theta + W_c = 4L \cos\theta + W_c $$
However, due to the parallel nature of the two sides, the relevant constraint for the symmetrical motion of our specific linkage is:
$$ W_{active} = 2L \cos\theta + W_c $$
The vertical height (compression) is governed by the vertical projection of the links:
$$ H_c = 2L \sin\theta $$
These equations define the relationship between the link angle \(\theta\) and the overall dimensions of the bionic robot during its energy storage phase.
Locking-Release Mechanism: The Incomplete Gear and Cable Drive
A major challenge in jumping robots is the controlled release of stored energy. Complex latches or solenoid triggers add weight and control complexity. We propose an elegantly simple solution integrated with the drive system.
- Drive Train: The motor drives a primary pinion. This pinion meshes with a secondary gear on whose shaft a winding drum (winch) is attached.
- Energy Loading: A high-strength polymer cable is wound around the drum. The free end of this cable is attached to the six-bar linkage’s moving connector. When the motor turns, the winch winds the cable, pulling the connector and thereby driving the six-bar linkage from its extended state to its compressed state, stretching the springs.
- The Incomplete Gear as a Trigger: The primary pinion is an incomplete gear—a standard gear with a section of its teeth removed. During the winding phase, the toothed section of the primary pinion meshes with the secondary gear, providing the necessary torque to wind the cable and compress the springs against increasing resistance.
- Automatic Release: As the motor continues to rotate, the pinion eventually rotates to its toothless sector. At this precise moment, the meshing disengages. The secondary gear and winch are now free to rotate independently of the motor. The stretched springs immediately contract, pulling the cable back and causing the winch and secondary gear to spin rapidly in the reverse direction. This rapid unwinding releases the constraint on the linkage, allowing the stored elastic energy to be converted into kinetic energy, launching the bionic robot.
This mechanism is highly efficient. The motor only needs to run continuously in one direction. It automatically loads energy until a mechanical “soft limit” (the end of the tooth sector) is reached, at which point release is instantaneous and autonomous. The cable transmission eliminates the need for rigid guides or slides, reducing friction and part count.
| Subsystem | Key Components | Primary Function | Innovative Feature |
|---|---|---|---|
| Chassis & Forelegs | Main plate, hinged forelegs | Provide structure, stabilize posture | Passive, adaptive foreleg angle for landing |
| Drive Unit | Geared DC motor (N20 type) | Provide mechanical power for energy loading | Compact, high-torque, low-speed output |
| Six-Bar Mechanism | Thigh/Shank links, sync gears, tension springs | Store elastic energy, coordinate leg extension | 1-DOF linkage for deterministic motion; Topology-optimized links |
| Locking-Release | Incomplete pinion, secondary gear, winch, cable | Control energy loading and trigger release | Automatic release via missing gear teeth; Low-friction cable drive |
Mechanical Design and Kinematic Analysis
Operational Sequence
The jumping cycle of this bionic robot is a periodic process with two main phases: the Energy Loading Phase and the Launch Phase.
1. Initial State & Energy Loading Phase: The robot starts with the six-bar linkage extended, springs relaxed, and the incomplete pinion’s toothless sector facing the secondary gear (no meshing). When the motor is activated, the pinion rotates. Once the toothed sector engages the secondary gear, the winch begins to wind the cable. The cable pulls the connector on the linkage, causing the thigh and shank links to fold. The synchronized gears ensure both legs move together. This folding action stretches the two tension springs, storing elastic potential energy \(E_s\). Concurrently, the entire robot body is pulled downwards and backwards, with the hind feet firmly on the ground and the forelegs tilting. The system reaches a fully compressed, high-energy state.
2. Trigger and Launch Phase: The motor continues running. The pinion rotates until its toothed sector completely passes the secondary gear. The meshing disengages the instant the toothless sector arrives. The secondary gear and winch are now mechanically decoupled from the motor. The potential energy in the springs is suddenly converted into kinetic energy:
$$ E_s = \frac{1}{2} k (\Delta x)^2 \times 2 = k (\Delta x)^2 $$
where \(k\) is the spring constant of each spring and \(\Delta x\) is the extension length of each spring. This energy accelerates the robot’s mass \(m\) upward and forward. The linkage rapidly extends, pushing the body away from the ground. The feet lose contact, and the bionic robot enters a ballistic flight trajectory.
3. Reset and Cycle Continuation: After the jump, the robot lands. If the motor is still powered, the pinion will complete its revolution. The toothed sector will re-engage with the secondary gear, and the winding process begins anew, resetting the system for the next jump. This allows for continuous, periodic jumping.
Key Parameter Calculations and Design Sizing
To maintain a biomimetic scale, the target overall size of the bionic robot was set at approximately 10 cm x 10 cm x 10 cm. The following key parameters were derived from kinematic constraints and performance goals.
1. Six-Bar Linkage Dimensions: Given a target compressed height \(H_c \approx 75 \text{ mm}\) and an active width \(W_{active} \approx 100 \text{ mm}\) with a gear center distance \(W_c = 15 \text{ mm}\), we solve for link length \(L\) and compression angle \(\theta\).
From the constraint equations:
$$ 100 = 2L \cos\theta + 15 $$
$$ 75 = 2L \sin\theta $$
Solving these yields:
$$ L = \frac{\sqrt{(100-15)^2 + 75^2}}{2} \approx \frac{\sqrt{7225 + 5625}}{2} = \frac{\sqrt{12850}}{2} \approx 56.7 \text{ mm} $$
$$ \theta = \arctan\left(\frac{75}{85}\right) \approx 41.4^\circ $$
For practical manufacturing and to avoid link interference, the link length was finalized at \(L = 42 \text{ mm}\). This changes the operational angle \(\theta\) during full compression, which is recalculated based on the chosen spring travel.
2. Spring and Cable Specification: The design calls for a cable pull distance (winch payout) of \(\Delta l_{cable} = 30 \text{ mm}\) to achieve full compression. High-manganese steel tension springs were selected with a free length of 50 mm and a spring constant \(k\). The required spring extension \(\Delta x\) is slightly larger than the cable pull due to linkage geometry, approximately 35 mm. The total energy stored is \(E_s = 2 \times \frac{1}{2} k (\Delta x)^2 = k (\Delta x)^2\).
3. Gear and Winch Design: The gears must withstand high impulsive loads during release. Standard module \(m=1\) gears were chosen. The incomplete pinion has 24 teeth, with 3 teeth removed to create the release sector. The secondary gear has 24 full teeth.
The winch must wind 30 mm of cable. For smooth release and to avoid excessive layers, the winch diameter \(d\) is calculated based on the required rotation. The release occurs when the pinion rotates through the angle corresponding to its 21 meshing teeth. The secondary gear and winch rotate a proportional amount:
$$ \text{Winch Rotation} = \frac{21}{24} \times 2\pi \text{ rad} = \frac{7}{8} \times 2\pi \text{ rad} $$
The circumference wound is \(\Delta l_{cable} = \pi d \times \frac{7}{8}\). Therefore:
$$ d = \frac{\Delta l_{cable} \times 8}{7\pi} = \frac{30 \times 8}{7\pi} \approx 10.9 \text{ mm} $$
A winch diameter of 11 mm was selected.
| Parameter | Symbol | Value | Unit |
|---|---|---|---|
| Target Robot Size | – | ~100 x 100 x 100 | mm |
| Thigh/Shank Link Length | \(L\) | 42 | mm |
| Gear Center Distance | \(W_c\) | 15 | mm |
| Cable Pull / Winch Stroke | \(\Delta l_{cable}\) | 30 | mm |
| Spring Extension (approx.) | \(\Delta x\) | 35 | mm |
| Gear Module | \(m\) | 1 | mm |
| Pinion Teeth (Total/Removed) | \(z_p\) | 24 / 3 | – |
| Secondary Gear Teeth | \(z_s\) | 24 | – |
| Calculated Winch Diameter | \(d\) | 11 | mm |
Dynamic Modeling and Performance Estimation
A simplified dynamic model provides insight into the expected performance of the bionic robot. We model the launch as an instantaneous conversion of spring energy into the kinetic energy of the robot’s center of mass (COM), followed by ballistic motion.
Take-off Velocity: Assuming negligible energy loss in the linkage and release mechanism (an ideal case), the stored spring energy equals the initial kinetic energy:
$$ E_s = k (\Delta x)^2 = \frac{1}{2} m v_{to}^2 $$
$$ v_{to} = \sqrt{\frac{2k (\Delta x)^2}{m}} $$
where \(v_{to}\) is the magnitude of the take-off velocity.
Jump Distance and Height: The take-off angle \(\phi\) is determined by the orientation of the force vector from the extending legs at the moment of ground contact loss. For a simplified analysis, we assume an optimal angle of \(\phi = 45^\circ\). The components of take-off velocity are:
$$ v_{x0} = v_{to} \cos\phi, \quad v_{y0} = v_{to} \sin\phi $$
Using projectile motion equations, the theoretical maximum range \(R\) and maximum height \(H\) are:
$$ R = \frac{v_{to}^2 \sin(2\phi)}{g} = \frac{v_{to}^2}{g} \quad \text{(for } \phi=45^\circ\text{)} $$
$$ H = \frac{(v_{to} \sin\phi)^2}{2g} = \frac{v_{to}^2}{4g} \quad \text{(for } \phi=45^\circ\text{)} $$
where \(g\) is the acceleration due to gravity.
Real-World Factors: In practice, efficiency \(\eta\) (less than 1) must account for friction, incomplete energy transfer, and non-ideal launch angle. The effective take-off velocity becomes \(v_{to\_eff} = \sqrt{\frac{2 \eta k (\Delta x)^2}{m}}\). For a bionic robot with mass \(m = 0.15 \text{ kg}\), spring constant \(k = 50 \text{ N/m}\), \(\Delta x = 0.035 \text{ m}\), and estimated efficiency \(\eta = 0.7\):
$$ v_{to\_eff} = \sqrt{\frac{2 \times 0.7 \times 50 \times (0.035)^2}{0.15}} \approx \sqrt{0.0572} \approx 0.24 \text{ m/s} $$
$$ R_{est} = \frac{(0.24)^2}{9.81} \approx 0.0059 \text{ m} \quad (5.9 \text{ mm}) $$
This simplified calculation yields an unexpectedly low result, highlighting a critical flaw in the initial parameter assumption. To achieve a jump of, say, 0.3 m (3x body length), the required take-off velocity would be \(v_{to} = \sqrt{Rg} = \sqrt{0.3 \times 9.81} \approx 1.71 \text{ m/s}\). Plugging this back into the energy equation reveals the necessary spring energy:
$$ E_s = \frac{1}{2} m v_{to}^2 = 0.5 \times 0.15 \times (1.71)^2 \approx 0.22 \text{ J} $$
For two springs, each must provide 0.11 J. Given \(\Delta x = 0.035 \text{ m}\), the required spring constant is \(k = \frac{2 \times 0.11}{(\Delta x)^2} = \frac{0.22}{0.001225} \approx 180 \text{ N/m}\). This indicates that stiffer springs or a longer draw distance \(\Delta x\) are essential for a high-performing bionic robot. This analysis is a vital part of the iterative design process.
| Parameter | Symbol | Initial Calculation | Revised Target | Unit |
|---|---|---|---|---|
| Robot Mass | \(m\) | 0.15 | 0.15 | kg |
| Spring Constant (each) | \(k\) | 50 | 180 | N/m |
| Spring Extension | \(\Delta x\) | 0.035 | 0.035 | m |
| System Efficiency | \(\eta\) | 0.7 | 0.7 | – |
| Effective Take-off Velocity | \(v_{to\_eff}\) | 0.24 | 1.71 | m/s |
| Theoretical Jump Range | \(R\) | ~0.006 | ~0.3 | m |
Control and Electronic Systems
The control system for this bionic robot is deliberately minimalist, aligning with the mechanical simplicity of the design. Its primary function is to activate the drive motor to initiate the energy-loading cycle. No active sensing or real-time feedback is required for basic periodic jumping.
Hardware Selection and Integration
- Actuator: A precision N20-type geared DC motor is used. This motor provides high torque at low speed, which is ideal for winding the cable against the increasing force of the springs. Its compact size and metal gearbox ensure durability and reliability for repeated cycling.
- Power Source: A lightweight 3.7V lithium polymer (Li-Po) battery with a capacity of 500 mAh supplies power. This battery offers a high energy density suitable for the small form factor of the bionic robot.
- Control Module: A compact integrated circuit board combines a radio receiver and a motor driver. The receiver operates on a common 2.4 GHz frequency band, offering a control range sufficient for demonstration (>20 meters). The motor driver is an H-bridge circuit that allows directional control of the motor with simple logic signals from the receiver.
- User Interface: A simple handheld remote transmitter with three buttons (Forward, Stop) is used for wireless command. Pressing “Forward” sends a signal to the receiver, which commands the motor driver to apply voltage to the motor, initiating the winding process. The “Stop” signal cuts power to the motor.
Circuit Configuration and Operational Logic
The wiring is straightforward. The battery connects to the power input terminals of the control board. The output terminals of the board’s motor driver are connected directly to the N20 motor. The system logic is as follows:
- Idle State: Robot on ground, springs relaxed, incomplete gear sector disengaged. Motor is off.
- Jump Command: User presses and holds the “Forward” button on the remote.
- Loading Sequence: The control board powers the motor. The motor rotates, the incomplete gear’s toothed section engages, and the winch winds the cable, compressing the linkage and storing energy. The user continues to hold the button.
- Autonomous Release: The pinion rotates into its toothless sector, mechanically disengaging. The springs release, causing the jump. The motor is still powered but is now disengaged from the winch and free-wheels.
- Cycle Reset: After the robot lands, the pinion, still rotating, will eventually bring its toothed sector back into engagement with the secondary gear. The winding process automatically begins again, resetting the bionic robot for another jump. The user can release the button to stop at any time.
This open-loop control scheme is remarkably effective for the designed mechanism, demonstrating how clever mechanical design can reduce electronic and control complexity.
Conclusion and Future Perspectives
This study presents the complete design rationale for a novel bionic robot capable of explosive jumping, inspired by the biomechanics of the frog. The core innovation is the integration of a constrained six-bar linkage as the primary energy storage and leg extension mechanism. This design elegantly transforms linear spring deflection into a synchronized, biomimetic leg motion while maintaining a single degree of freedom for reliability. Coupled with the incomplete gear and cable-driven locking-release system, the bionic robot achieves autonomous, repeatable jumping with minimal electronic control intervention. The use of topology optimization on critical load-bearing links further enhances performance by minimizing weight without compromising strength.
The kinematic and dynamic analyses provide a framework for understanding and predicting the robot’s performance. They also reveal critical design trade-offs, such as the necessity for high spring stiffness or long stroke lengths to achieve substantial jumping distances—a key consideration for the next iteration of this bionic robot. The proposed mechanical architecture successfully decouples the slow, high-torque energy-loading process (handled by the motor) from the fast, high-power energy-release process (handled by the springs), which is a fundamental principle in many biological jumpers.
Future work will proceed in several directions. Firstly, a physical prototype will be fabricated using 3D printing and composite materials to validate the design and refine the dynamic models with empirical data. Secondly, the current design can be enhanced by incorporating a mechanism to adjust the take-off angle \(\phi\), perhaps through a movable foreleg or chassis pivot, enabling controlled jump distance versus height. Thirdly, integrating a small inertial measurement unit (IMU) and a programmable microcontroller could allow for mid-air posture adjustment or adaptive jumping based on simple environmental sensing, moving from an open-loop to a closed-loop bionic robot. Finally, exploring different linkage topologies or spring configurations could yield variants optimized for specific tasks, such as vertical jumping for height or multi-directional jumping. This research establishes a robust and scalable platform for advancing the capabilities of bio-inspired jumping robots, contributing to their potential application in fields such as environmental monitoring, search and rescue, and planetary exploration.