Design and Experimental Validation of the Separation and Deployment System for the CHERI Lunar Micro-Rovers
Authors: Ahmet Mustafa Kangal, Oğuzhan Şahan, Teoman Efe Aycan, Görkem Egemen Güloğlu, Halil Ersin Söken, Ali Emre Turgut, Uluç Saranlı, Mustafa Mert Ankaralı
Organizations: Robotics Graduate Program, Graduate School of Natural and Applied Sciences, Middle East Technical University (METU), Ankara, Türkiye · Department of Aerospace Engineering, Middle East Technical University (METU), Ankara, Türkiye
Reliable separation mechanisms are critical for planetary micro-rovers, which must withstand launch vibrations and then eject the vehicle clear of the lander. This paper presents the design, optimization, and testing of the separation system of the CHERI (Challenging Environment Exploration Rovers for Intelligence) lunar micro-rovers. Sized for a 5 kg vehicle, the assembly comprises an aluminum ground plate, a central non-explosive hold-down and release mechanism, and an asymmetric four-spring ejection array. Initial separation trials revealed a velocity deficit caused by energy absorption in the flexible wheel-legs and chassis during the ejection stroke. This loss was represented as equivalent chassis damping in a multi-body dynamic model, and the springs were re-optimized accordingly. Early hardware iterations failed vibration testing, first through spring jamming in the guide housings and then through a 34 Hz fundamental mode of the flat ground plate. A density-based topology optimization raised the fundamental frequency to 84 Hz, above the 70 Hz requirement of the lander, within a 1.6 kg mass budget. The final configuration passed acceptance-level sinusoidal vibration, random vibration, and shock testing without structural degradation. Separation tests with a weight-matched one-sixth mass model, tracked by optical motion capture, confirmed clearance of the lander and reproduced the simulated free flight to within the run-to-run variation of the measured ejection velocity. These tests further showed that landing outcome is dominated by touchdown and settling dynamics, indicating that the residual pitch rate is the main controllable design parameter and confirming self-righting recovery as an integral element of the deployment architecture.
Figures & tables
Figure 1: Concept illustration of the CHERI micro-rovers (AI-generated image, shown for illustration only).
Figure 2: The CHERI micro-rover platform: (a) CAD render of the rover; (b) orthographic technical drawing indicating key structural dimensions and the hold-down interface; (c) the manufactured Structural Model (SM) used for mechanical verification and environmental testing.
Figure 3: Free-fall deployment kinematics showing the stowed vertical mounting height ( H=1.7m ) and the minimum dynamic horizontal clearance ( Dclearance≥0.7m ) required to clear the host lander’s landing pads.
Figure 4: Labeled exploded assembly view of the CHERI micro-rover separation interface, detailing the key physical components and sequence of structural integration: (0) lander surface, (1) isolation pads, (2) ground plate, (3) mounting screws, (4) separation springs, (5) separation nut, and (6) CHERI Rover.
Figure 5: First technological era using micro shape memory alloy (SMA) release nuts (Iterations 1–3): (a–c) isometric CAD models; (d–f) manufactured hardware prototypes.
Figure 6: Second technological era using a fracture-type separation nut 22 (Iterations 4–6): (a–c) isometric CAD models; (d–f) corresponding physical hardware.
Iter.
Actuator Type
Spring Config.
Mass (kg)
1st Nat. Freq.
Critical Failure Mode
Status
1
2 × Release Nut
None (Gravity)
0.85
NA
Actuator synchronization; no ejection force
NA
2
1 × Release Nut
2 × Lateral
0.92
NA
Insufficient clearance (clips lander footpad)
NA
3
1 × Release Nut
2 × Vertical
1.05
NA
Inadequacy under cyclic test conditions
NA
4
1 × Separation Nut
4 × Asymmetric
1.28
51 Hz
Spring stiction; spring jam in Z-axis random vib.
Failed
5
1 × Separation Nut
4 × Asym. (Opt. Guides)
1.48
34 Hz
First mode below lander requirement (70 Hz )
Failed
6
1 × Separation Nut
4 × Asym. (Opt. Guides)
1.54
84 Hz
None (SIMP topology-optimized plate)
Passed
Table 1: System-level engineering comparison of the six evolutionary separation system generations.
Figure 7: Chronological evolution of the separation spring mechanisms, detailing CAD cross-sections (top row) and corresponding physical hardware (bottom row). Left (Phase A): the initial enclosed COTS-based baseline. Middle (Phase B): the first custom iteration, whose unoptimized guide allowed lateral buckling and sliding stiction during release. Right (Phase C): the final configuration with the re-engineered guide bushing and rod. Key components are annotated: (1) spring hat, (2) helical spring, (3) guide bushing, and (4) central guide rod. The re-engineered guide bushing (3) and rod (4) ensure a stable, linear stroke without buckling or stiction.
Figure 8: Two-stage parametric spring optimization: (a) increasing the spring rate by +53% while keeping the energy delivered per spring over the separation stroke constant at 680.3mJ ; (b) increasing the free length from 82.0mm (Proposed) to 85.0mm (Optimum), which raises the spring force at the release point ( 75.0mm ) from ≈9.8N to ≈12.1N with a similar stowed force.
Config.
Free length
Rate
Coils
Rate incr.
Lf (mm)
k (N/mm)
nt
(%)
Original
92.4
0.910
16.0
–
Proposed (82)
82.0
1.396
11.1
+53.4%
Optimized (85)
85.0
1.209
12.5
+32.9%
Table 2: Geometric and physical comparison of the evolutionary compression spring iterations.
Figure 9: Finite element modal analysis comparison showing the first fundamental mode shapes of (a) the flat fifth-generation ground plate exhibiting severe central flexing at 34Hz , and (b) the topology-optimized sixth-generation ground plate with stiffening ribs raising the natural frequency to 84Hz .
Mode
Deformation Shape
Pre-Opt.
Post-Opt.
Status
No.
Iter. 5 (Hz)
Iter. 6 (Hz)
( ≥70Hz )
1
Longitudinal Torsion
33.6
84.0
Passed
2
Out-of-plane Bending
55.4
116.9
Passed
3
Lateral Torsion
59.8
107.2
Passed
Table 3: First three natural frequencies of the stowed separation assembly before and after topology optimization (FEA). Modes are matched by deformation shape.
Figure 10: Structural test and calibration setups at the TÜBİTAK UZAY satellite testing facility: (a) the stowed separation assembly mounted on the electrodynamic shaker table for vibration testing in three orthogonal axes, (b) the pneumatic shock test setup, in which an aluminum projectile strikes a resonant plate, and (c) the non-functional dummy model used as a mass simulator.
Quantity
Scaling law
Ratio (Earth/Moon)
Weight
f(g0/gMoon)
1.01
Downrange at touchdown
f−1/2(gMoon/g0)1/2
1.00
Attitude at touchdown
f−1/2(gMoon/g0)1/2
1.00
Translational kinetic energy
f⋅f−1
1.00
Rotational kinetic energy
f⋅f−1
1.00
Separation velocity
f−1/2
2.45
Table 4: Ratio of the terrestrial 1/6 mass test to lunar flight conditions under weight-matched Froude scaling, with f=msim/mflight=1/6 and g0/gMoon=6.06 . Quantities with unit ratio transfer exactly; the remainder do not.
Figure 11: Physical experimental launch test setup, showing the test arena with wall-mounted optical motion-capture cameras and the landing sandbox (left), and the 1/6 mass model stowed and preloaded on the ground plate (right).
Figure 12: Terrestrial test rig at three representative slope conditions used in the campaign of Section 6.3 : (left to right) 0∘ , +10∘ and +20∘ . Steeper and clockwise ( −20∘ and beyond) conditions are reported in Table 5 but not shown here, as they were tested with fewer repetitions after early overturns (Section 6.3 ).
Slope (deg)
Lunar sim.
Earth sim.
Earth exp.
−45
7/10
0/10
0/5
−30
10/10
2/10
0/5
−20
10/10
0/10
0/5
−10
10/10
8/10
3/10
0
10/10
8/10
4/10
+10
10/10
9/10
5/10
Table 5: Separation and landing outcomes (successes/runs) over the slope range. Lunar simulation: 5.0kg , g=1.62m/s2 . Terrestrial simulation and experiment: both 0.833kg , g=9.81m/s2 . Positive slopes are counter-clockwise, negative clockwise. Cells with 5 runs were truncated after repeated failure to conserve separation-nut actuations.
Figure 13: Motion-capture trajectory of the 1/6 mass model during separation at +20∘ slope (right), compared with the corresponding terrestrial simulation (left). Gray markers denote interpolated positions where tracking data were unavailable.
Figure 14: Motion-capture trajectory of the 1/6 mass model during separation at −20∘ slope (right), compared with the corresponding terrestrial simulation (left).
Figure 15: Simulated landing outcomes on a −20∘ slope. Left: lunar conditions, full flight mass, successful wheels-first landing. Right: terrestrial conditions, 1/6 mass model, overturned. The free-flight trajectories are geometrically identical; the divergence occurs entirely during touchdown and settling.
Lunar micro-rovers under 50 kg are a rapidly growing vehicle class, yet open, benchmarked tools for running optimization across design trades remain scarce. Rover sizing is a highly coupled problem with a feedback loop specific to surface vehicles: a heavier rover sinks deeper into loose regolith, deeper sinkage costs more power, more power requires a larger array, and the array adds mass. We present RoverDevKit, a tradespace exploration toolkit with an open evaluator that couples a closed-form wheel-soil model with bottom-up mass, solar-power, thermal-survival, and traverse models. It is fast enough to serve as the fitness function of a multi-objective genetic algorithm (NSGA-II): one search of 3000 mission evaluations takes under a minute on a single laptop core. We use it to map the trades among distance traveled, vehicle mass, and slope capability on mare, polar, highland, and crater-rim missions. Across those missions the limiting factor changes: energy storage at high latitude, slope traction on loose highland regolith, and traverse range on mare and crater-rim terrain. The wheel-soil model matches measured single-wheel drawbar pull on two independent datasets to within the 20-30% error typically reported for this class of model, and the mass model predicts published 5-50 kg rover masses to 10.5% median absolute error. A check against five published in-class designs places them near the designs the optimizer selects. Re-running the search after applying that wheel-soil comparison error leaves these conclusions unchanged.
Legged robots for lunar exploration must be designed for low gravity, where reduced contact forces and long flight phases make leg morphology and controller choice interdependent. We study this for a planar monoped in lunar gravity (1.625 m/s2) using a bi-level co-design framework: a differential evolution outer loop searches over link lengths, and each candidate is evaluated on a vertical jump by either a quadratic program based task-space controller (QP-TSC) or a direct-collocation based trajectory optimization. The two controllers yield different optimal link lengths, and trajectory optimization outperforms the QP-TSC across most of the design space, with 17% lower Cost of Transport (COT) at their respective optima. We then extend to a 2D jump, where trajectory optimization with centroidal angular momentum regulation produces stable in-place and forward jumps, and an LQR controller absorbs impact and stabilizes the robot after touchdown. This is demonstrated on a baseline morphology with both links 0.3 m long.
Sanjay Ramkumar, Suraj Kumar
Dept. of Aerospace Engineering, Indian Institute of Space Science and Technology. · Controls and Digital Area, U R Rao Satellite Center.
Future planetary rovers will traverse steep, deformable terrain where slip, sinkage, and terrain deformation can threaten mobility. Although simulation can capture some wheel-soil effects, unmodeled deformable-terrain phenomena create sim-to-real gaps requiring deployment-time adaptation. Yet, rover operations leave little room for extensive trial-and-error. We introduce TERRA (Targeted Exploration for Reward-Relevant Adaptation), an online adaptation framework that efficiently seeks deployment data exposing task-relevant model errors and corrects them. TERRA adapts a simulation-trained world model with a reward-weighted prediction loss and trains a loss predictor to estimate where future performance-relevant errors are likely to occur, using this signal to guide exploration during MPC. In simulation, TERRA reaches approximately 90% slope-climbing success within five trials under severe out-of-distribution pretraining conditions, outperforming exploration baselines in reward, completion time, and energy to first success. On the ERNEST rover, TERRA recovers coordinated active-suspension locomotion on a deformable Mars-simulant slope within three to five real-world trials. Project website: https://terra-explore.github.io
Jacob Levy, Arthur Bouton, Tristan D. Hassler +2
Department of Aerospace Engineering, University of Texas at Austin · Jet Propulsion Laboratory, California Institute of Technology