From the Atacama to Mars: The Tumbleweed Rover’s Analog Campaign to Validate Surface Operations and Scientific Return

Introduction

Exploration of the Martian surface is currently limited by the low spatial coverage achievable with conventional rover architectures. To address this limitation, Tumbleweed Mars has developed its next-generation tumbleweed-bioinspired rover platform into a mission concept consisting of a swarm of wind-driven, spheroidal rovers designed to enable large-scale, in-situ exploration of Mars with minimal energy requirements [1].

The Tumbleweed Mission represents a novel paradigm for scalable large-scale Mars exploration using a fleet of tumbleweed rovers. This abstract summarises the initial results from the April 2026 Atacama Mars-analog Campaign, focused on validating the TRL-6 mobility and surface operations of the rover.

The Tumbleweed Mission

The Tumbleweed platform builds on the pioneering work performed in the early 21st century concerning the design and testing of precursor Tumbleweed rovers [2]. Extending on previous work, Tumbleweed Mars has extensively researched how upgraded box-kite-styled Tumbleweed rovers could enable the investigation of Mars’ atmosphere and ionizing radiation environment [3], search for biosignatures [4], constrain geological history and modern geomorphology [5] and identify future Landing Sites for Human missions to Mars [6]. Thus, the proposed Tumbleweed Mission (figure 1) will accelerate the pace of surface exploration and data acquisition on Mars, and fill the capability gap that exists between orbital and conventional rover platforms.

Figure 1 - Proposed Mission Architecture: mid-air deployment (1), landing (2), rolling towards the equator (3) and stationary phase (4).

 

The Atacama Campaign

In 2025, Tumbleweed Mars achieved validation of the current box-kite rover spheroidal rover design through a wind tunnel campaign in Aarhus University Planetary Environment Facility supported by Europlanet’s Transnational Access Program [7]. To further validate Martian surface operations, as well as high-quality scientific return during dynamic motions, an analog campaign in the Atacama desert was organised in collaboration with UC Chile, and through funding with TU Delft FAST grant. 

Thus, a 12-day Tumbleweed Martian analog campaign was performed between 23rd March and 3rd April 2026 around the research station Oasis de Niebla Alto Patache. To this end, Tumbleweed Mars developed a 1.5-meter prototype (figure 2), equipped with modular sensors to collect wind, temperature, pressure, and other environmental data. 

Figure 2 - Tumbleweed rover prototype during the Atacama desert testing campaign (April 2026). 

The main research objectives of the campaign were the following:

• Validating long-term and long-distance (12 km) autonomous rover operations in Martian-analog terrain, taking the TRL of the surface operations and mobility subsystems to 6.
• Demonstrating analogous operations of the scientific suite on board a rolling Tumbleweed rover prototype.

The Atacama Desert is by far the driest and oldest desert on Earth and is ideal as a Mars analog, due to the similar surface properties and geomorphological features. Most importantly, for Tumbleweed rovers, which are passively driven by the wind, predictable and regular wind conditions are required. During the southern autumn period in the Atacama desert, local winds maintain a consistent diurnal cycle, with wind in the range of 3-4 m/s during the day and decreasing to 1-2 m/s during the night, which is qualitatively consistent with the diurnal Mars wind regimes. 

Results

During the campaign, the rover successfully achieved a maximum traverse of approximately 12 km. Experimental results established a wind speed threshold for movement on flat ground at 2.6–2.8 m/s, scaling to an estimated 12 m/s in the Martian environment for a 5-meter diameter rover. A key achievement was the autonomous navigation of inclines; the rover successfully climbed slopes of 10 to 15 degrees in gusting conditions up to 5 m/s.

Figure 3 - A 1.5 km traverse guided by a northeastern wind direction against a slope and harsh terrain, with temperature data illustrated along the traverse. 

Structural analysis revealed a critical trade-off between the flexibility required for landing/transit and the rigidity needed for surface operations. The implementation of equidistant clamps on the outer structure significantly reduced side-to-side swaying and increased operational stability. While the system demonstrated high resilience, with single rod breakages failing to impact motion, adjacent failures or extreme terrain traps (e.g., narrow riverbeds or boulders) were identified as primary failure modes. These results provide a foundational dataset for refining autonomous pathfinding and structural design for future Martian deployment. Nevertheless, the prototype showed sustained navigation capability along harsh terrain, upslope, through passive wind-based propulsion (figure 3).

Figure 4 - Dynamic shot from the HQ camera with an outer structure rod visible. 

Onboard instrumentation, including HQ and Wide-Angle (WA) cameras, captured extensive environmental data during a 12 km traverse. Initial analysis of the HQ imagery (figure 4) suggests sufficient resolution for grain-size analysis and geological history mapping, verified against 300g physical regolith samples collected at test sites. However, the campaign highlighted the need to couple shutter speed and gain to the light flux sensor and IMU to mitigate motion-induced artifacts. Environmental sensors successfully recorded temeprature, pressure, light intensity, UV index, and humidity along traverses (figure 3, for GPS and temperature along a traverse).  Further, the "nutation" or wobbling caused by mass distribution offsets will be modeled in Unity simulations to optimize future instrument stability.

In conclusion, the Atacama campaign provided strong evidence that the Tumbleweed rover concept can support long-range, low-energy surface exploration in Mars-analog terrain. The Tumbleweed rover demonstrated robust mobility, autonomous slope-climbing, and stable scientific data collection over a 12 km traverse, while also revealing key design constraints in structure, sensing, and motion stability. The results highlight the value of coupling environmental measurements with rover dynamics to improve future scientific inference, navigation, and payload performance. This marks an important step towards validating a scalable swarm-based mission architecture for distributed, in-situ Mars exploration.

 

References

[1] Kingsnorth, J. et al., EGU Gen. Assem., EGU24-20149, 2024.

[2] Behar, A. et al., IEEE Aero. Conf., Vol. 1, 2004.

[3] Shanbhag, A. et al., IAC-23, A1.5, 2023.

[4] Shanbhag, A. et al., IAC-23, A1.6, 2023.

[5] Kingsnorth, J. et al., IAC-23, A3.IP, 2023.

[6] Tjokrosetio, D. et al., IAC-23, E5, 2023.

[7] Kingsnorth, J. et al., EPSC-DPS, 2025-1775, 2025.