A concept-level exploration of the NASA Rock and Roll Challenge: how an AI agent, guided by TRIZ Consulting tools, transformed a complex technical challenge into contradictions, inventive principles and reasoned design directions.

AI and TRIZ on the NASA Rock and Roll Challenge: from contradiction to concepts

The NASA Rock and Roll Challenge is one of those technical problems that does not simply ask for “a better solution”.

It asks for something more difficult: holding two opposite properties together.

Here the link to the full description of the challenge: https://www.herox.com/NASARockandRoll

The challenge concerns the development of a new generation of lunar rover wheels: lightweight, durable, able to deform in a controlled way to absorb impacts, but also rigid enough to transmit traction and support high loads.

All of this on lunar regolith.

Regolith is the layer of dust, rock fragments and granular material that covers the surface of the Moon. It is not just “sand”: it can be abrasive, sharp and highly problematic for any mechanical component in contact with the ground.

The challenge addressed using TRIZ Consulting AI tools involves severe requirements:

• mass below 2.3 kg
• speed up to 24 km/h
• operational life of 1,000 km
• dynamic loads up to 4 times the nominal static load
• operation across an extreme thermal range around from -240°C to +120°C.

But the interesting point, for anyone working on innovation, is not simply designing a wheel.

This is not a normal improvement request.

The wheel must be flexible under impact and rigid when transmitting traction.
It is a contradiction.
And this is exactly the type of situation where TRIZ becomes useful.

Want to try the same approach?

Take one of your technical challenges and try framing it with TRIZ Consulting’s AI tools.

Look for a contradiction: what do you want to improve? What gets worse when you try?

Why this challenge is a good test for AI + TRIZ

In previous articles in the Challenge to Clarity series, we used public challenges as test cases to show one precise point: AI becomes more useful when it is not used only to “generate ideas”, but when it is guided by a method.

In this new case, we applied TRIZ Consulting AI tools to the NASA Rock and Roll Challenge with a specific objective:

not to find “the final solution”, but to structure the problem in order to generate more reasoned design directions.

The challenge is interesting because it concentrates extreme constraints into a single system: low weight, durability, speed, abrasion resistance, impact absorption, traction on slopes, compatibility with rover architectures and operation in the lunar environment.

In other words: making the wheel “stronger” is not enough.

Making it stronger may increase weight, rigidity and localized stress.
Making it more flexible may improve impact absorption, but worsen traction, stability and durability.

The real problem is not a single performance parameter.

It is the conflict between performances.

The decisive step: turning the problem into a contradiction

A traditional approach could start immediately from solutions:

• use stronger materials;
• increase thickness;
• introduce an elastic structure;
• change the tread;
• add reinforcements;
• use more complex geometries.

All these directions are possible.

But they risk starting too early.

The TRIZ workflow starts from a different question:

which characteristic do we want to improve, and which characteristic worsens when we try to improve it?

In the case of the lunar wheel, the main technical contradiction can be formulated as follows:

• improve the structural flexibility of the wheel, meaning its ability to deform elastically and adapt to the terrain;
• without worsening its mechanical strength and structural integrity, meaning its ability to maintain geometry and performance after prolonged exposure to abrasive regolith.

The physical contradiction is even more direct:

the same wheel must be flexible and rigid.

Not in abstract terms.
Not as a theoretical elegance.
But because in some conditions it must yield, and in others it must resist.

It must yield locally when it encounters asperities and impacts.
It must remain globally stable when it transmits traction and supports the load.

This distinction changes the way we think.

The problem is no longer:

“Which material should we choose?”

It becomes:

“Where do we need flexibility? Where do we need rigidity? When do we need one? When do we need the other? Can we separate them in space, in time or according to operating conditions?”

This is where TRIZ reduces ambiguity.

What TRIZ Consulting AI tools did

In this experiment, we did not simply ask an AI model:

“Find ideas for an innovative lunar wheel.”

We did something different.

We gave an AI agent an operational prompt: read the challenge, interpret its constraints and use TRIZ Consulting tools to structure the problem and generate preliminary concepts.

This detail matters.

The solutions were not manually built by an expert during the process. They emerged from the interaction between the AI agent, the TRIZ workflow and the criteria of the challenge.

Human intervention was concentrated upstream, in defining the task, and downstream, in critically reviewing the output.

The workflow guided the analysis through a structured sequence: context, objectives, KPIs, problem, constraints, functional diagram, contradiction, concept generation and final ranking.

The challenge was framed as a new product development case in the aerospace / lunar mobility systems sector. The report then led to the generation of three main concepts: LunaMesh, PoliSegm and BioMech.

This sequence is the point.

There is a difference between obtaining a list of ideas and obtaining a reasoning map.

A list may look creative, but it is often dispersive.
A map helps explain why a design direction makes sense, which constraints it attempts to address and which risks remain open.

From TRIZ principles to concepts

The analysis highlighted several TRIZ principles that are particularly coherent with the problem.

From these principles, three main concepts emerged.

1. LunaMesh

A three-dimensional mesh wheel based on SMA material, built around the idea of a structure that deforms elastically and tends to recover its original geometry.

The conceptual advantage is clear: the wheel itself becomes the shock-absorbing system, without introducing additional active components.

2. PoliSegm

A modular segmented wheel with curved elements and elastic pantograph joints.

Each segment can partially adapt to the terrain profile, better separating the contact function from the structural function.

In the preliminary ranking of the report, this concept emerged as the most balanced in terms of desirability, feasibility, sustainability, effort and risk.

3. BioMech

A bio-inspired compliant mechanism, with a logic similar to an “arachnid leg”: a network of flexible elements that distributes the load and reduces dependence on traditional joints.

It is a more exploratory concept, interesting for its originality, but also riskier from a validation perspective.

The value is not “having found the right wheel”

This point needs to be stated precisely.

The generated concepts are not validated solutions.

They are not test results.
They are not engineering proposals ready for flight.
They do not demonstrate that a specific architecture meets NASA requirements.

They are a concept-level exploration.

Their value is different: they show how a TRIZ + AI workflow can transform a complex challenge into a set of reasoned design directions.

This is often the most underestimated step in innovation processes.

Many companies do not have a lack-of-ideas problem.

They have a dispersion problem.

Ideas arrive.
Constraints multiply.
Hypotheses overlap.
Discussions become circular.
Every business function sees a different risk.

At that point, generating more ideas may make the situation worse.

What is needed is structure.

Why this case matters beyond aerospace

The lunar wheel is an extreme case.

But the shape of the problem is very common.

A product needs to be:

• lighter but stronger;
• faster but more precise;
• cheaper but more reliable;
• simpler but higher-performing;
• more flexible but more controllable;
• more standardized but more customizable.

Almost every complex industrial project contains this type of contradiction.

The difference is that it is often treated as a compromise.

TRIZ treats it as working material.

When a contradiction is well formulated, the team stops asking only “how much do we need to sacrifice?” and starts asking:

how can we eliminate the need to sacrifice?

This does not automatically guarantee a solution.

But it changes the quality of the reasoning.

What this exploration shows

This concept-level exploration of the NASA Rock and Roll Challenge shows four things.

First: AI is more useful when it is channeled. Without a method, it tends to produce alternatives. With a structured workflow, it can help build a logical progression.

Second: TRIZ is particularly effective when the problem contains opposite properties. Here the wheel does not need to be “a little flexible and a little rigid”. It needs to separate two behaviors according to zone, function and operating condition.

Third: the generated concepts are useful because they make assumptions visible. Each direction carries benefits, risks and required tests.

Fourth: the value of the workflow is not to replace engineering. It is to prepare it better.

Before simulation.
Before prototyping.
Before testing.

There is an often invisible phase: clarifying the problem.

That is where much of the quality of the following work is decided.

Conclusion

The NASA Rock and Roll Challenge is interesting not only because it concerns lunar exploration.

It is interesting because it makes visible a dynamic found in many innovation projects:

the problem is not choosing between two opposite properties.
The problem is understanding how to separate them.

This is where TRIZ and AI can work together.

AI accelerates exploration, synthesis and generation.
TRIZ gives structure, direction and discipline to the reasoning.

The result is not a magic answer.

It is something more useful:
a clearer path for moving from an ambiguous challenge to concepts that can be discussed, evaluated and improved.

---

Series: Challenge to Clarity

Suggested slug: ai-triz-nasa-rock-and-roll-challenge

Meta description: A concept-level exploration of the NASA Rock and Roll Challenge: how TRIZ Consulting AI tools help reduce ambiguity, reveal contradictions and generate reasoned design directions.