Earlier this year, a robot completed the half marathon in Beijing in less than 40 minutes. It's slower than the human winner, who has played in the competition for more than an hour, but it's still an amazing feat. Many casual runners will be proud of that time. The robot has a pace of more than 13 miles (21 km).

But this does not charge at once. Along the way, the robot had to stop and replace its battery three times. This detail, while easily overlooked, illustrates a deeper challenge in robotics: energy.

Modern robots can move with incredible agility, mimic the movement of animals, and perform complex tasks with mechanical precision. In many ways, they are comparable to biology in coordination and efficiency. However, when it comes to endurance, robots are still lacking. They won't get tired of fatigue- they just run out of strength.

As a robotics researcher, I focus on energy systems and I looked at this challenge carefully. How can researchers provide robots with the lasting power of biological organisms - why are we still so far away from this goal? While most robotics on energy problems focus on better batteries, there is another possibility: robots that make food.

The robot moves well, but runs out of steam

Modern robots are very good at moving. Thanks to decades of research in biomechanics, motion control and drive, machines like Boston Dynamics and Atlas can walk, run and climb with agility, which once seemed out of reach. In some cases, their motors are more effective than animal muscles.

But endurance is another matter. For example, the spot can be fully charged for only 90 minutes. After that, it takes nearly an hour to charge. These runs are a transition from eight to 12 hours for human workers, or multi-day endurance for sled dogs.

The question is not how the robots move, but how they store energy. Today, most mobile robots use lithium-ion batteries, the same as smartphones and electric cars. These batteries are reliable and available, but their performance is improved at a slow pace: every year, the new lithium-ion batteries are 7% better than the previous generation. At this speed, it would take a full decade to just double the runtime of the robot.

[embed]https://www.youtube.com/watch?v=I44_ZBEWZ_W[/embed]

Animals store energy in fat, which is a dense energy-rich: nearly 9 kWh per kilogram. By comparison, this is a sled dog that totals about 68 kWh, similar to the energy of a fully charged Tesla model. By comparison, lithium-ion batteries store only a small portion, about 0.25 kWh per kilogram. Even with efficient motors, robots like Spot need to be dozens of times stronger than today's batteries to match the endurance of sled dogs.

And charging is not always an option. In disaster areas, remote control areas or long-term tasks, wall outlets or backup batteries may not be visible anywhere.

In some cases, robot designers can add more batteries. However, more batteries mean more weight, which increases the energy required to move. In highly mobile robots, there is a careful balance between payload, performance and endurance. For example, a battery already accounts for 16% of its weight, for example.

Some robots use solar panels, which, in theory, may prolong their runtime, especially for low-power tasks or in bright sunny environments. But in reality, solar power has little power compared to something that mobile robots need to walk, run or fly at actual speeds. That's why energy collection like solar panels is still a niche solution today, better suited for fixed or ultra-low power robots.

Why it matters

These are not just technical limitations. They define what a robot can do.

The rescue robot with a 45-minute battery may not last enough to complete the search. A farm robot that pauses and charges every hour is unable to harvest crops in time. Even in a warehouse or hospital, short time increases complexity and cost.

If robots are to play a meaningful role in society to assist older people, explore dangerous environments and work with humans, they need endurance to stay active for hours, not minutes.

New batteries such as lithium sulfur and metal air are chemically distributed as the way forward. These systems have much higher theoretical energy density than today's lithium-ion cells. Some methods seen in animal fat. When paired with an actuator that effectively converts electrical energy from a battery to a mechanical work, they can enable the robot to match or even exceed the endurance of animals with low body fat. But even these next-generation batteries have limitations. Many people have difficulty recharging, degrading over time or facing engineering barriers in the real world.

Fast charging can help reduce downtime. Some emerging batteries can be charged in minutes instead of hours. But there are trade-offs. Fast charging strain battery life, increases heat, and often requires a large amount of high-power charging infrastructure. Even with improvements, fast-charging robots still need to stop frequently. In an environment where grid power is not available, this does not solve the core problem of limited on-board energy. That's why researchers are exploring alternatives such as "refueling" robots with metal or chemical fuels (like animals eat) to bypass the limits of electrical charging.

Alternative: Robot metabolism

In nature, animals do not charge, they eat. Food is converted into energy through digestion, circulation and breathing. Fat stores energy, blood moves it, muscles use it. Future robots can follow similar blueprints and have anabolic metabolism.

Some researchers are building systems that allow robots to "digest" metal or chemical fuel and breathe oxygen. For example, synthetic gastric acid chemical reactors can convert high-energy materials such as aluminum into electrical energy.

This is based on many advances in robot autonomy, where robots can sense objects in the room and navigate to pick them up, but here they pick up energy.

Other researchers are developing fluid-based energy systems that circulate like blood. An early example, a robot fish, tripled its energy density by using a multifunctional fluid instead of a standard lithium-ion battery. This single design transfer provides a 16-year battery improvement, not through new chemical reactions, but through a more biologically inspired approach. These systems can allow robots to operate over longer periods, drawing more energy from the materials that store energy than today’s batteries.

In animals, energy systems can not only provide energy. The blood helps regulate temperature, deliver hormones, fight infections and repair wounds. Anabolic can do the same thing. Future robots may use circulation fluids to manage heat, or use stored or digested materials to heal themselves. Energy can be stored throughout the body, limbs, soft thin-layer components instead of the central battery pack.

This approach may result in a machine that is not only durable, but is more adaptable, resilient and lifelike.

Bottom line

Today’s robots can jump and sprint like animals, but they can’t go far.

Their bodies are fast, their minds are improving, but their energy systems are not catching up. If robots are going to work with people in a meaningful way, we need to give them more intelligence and agility. We need to give them endurance.