Emergent Cooperative Thermoregulation in Lunar Swarm Robotics

Anais Jeger; Harry Holt; Lennart Puck

cs.ROOct 22, 2026

Emergent Cooperative Thermoregulation in Lunar Swarm Robotics

Authors: Anais Jeger, Harry Holt, Lennart Puck

Organizations: Center for Bits and Atoms, MIT · Advanced Concepts Team, ESA · Planetary Robotics Laboratory, ESA

Abstract

Swarms of small robots are an attractive way to survey lunar caves and lava tubes. Missions such as NASA's CADRE will fly cooperating rovers for distributed mapping, but only in daylight. The 354 h lunar night drops the surface towards 95 K, and the radioisotope heaters that let single rovers survive it cannot be supplied at swarm scale. In this study we investigate whether a swarm can instead use its own members as thermal shielding. We train one policy, shared by all twenty robots, by reinforcement learning with an evolution-strategies estimator. The policy receives a single reward, once, at the end of the night: the mean temperature of the five coldest robots. Nothing in that reward refers to position, spacing, exposure or motion. Two behaviors nonetheless emerge. The swarm aggregates, and robots circulate between the sheltered core and the exposed rim, the same pair of behaviors with which emperor penguins survive the Antarctic winter. We measure the survival threshold, the smallest stored energy per robot at which every robot is still alive at sunrise. Across ten runs it is 155.6 +/- 1.3 pcm-h, where one pcm-h is 7.2 kJ of stored heat. That is 37% below a motionless huddle and within 13% of a physics-based lower bound, and at a budget where the motionless huddle loses six robots, every robot survives. The behavior needs no shared spatial reference and no robot-robot contact, and transfers from twenty robots to forty without retraining.

Figures & tables

Panel of Figure 1 — Fig. 1: Twenty simulated robots during the $354\text{\,}\mathrm{h}$ lunar night, from one rollout of the learned policy. Ring color is each robot’s sky exposure $\chi$ on the scale of Fig. 3 . Top: the scattered start at hour 0. Every robot is red: with no neighbor close enough to block it, each one radiates to the full sky. Bottom: hour 73, after the swarm has gathered. The interior is blue, shielded by the bodies around it, and only the rim stays warm-colored; all twenty are alive.

Parameter	Value	Parameter	Value
$N$	20	$T_{\mathrm{set}}$	$273.15\text{\,}\mathrm{K}$
$R$	$0.085\text{\,}\mathrm{m}$	$T_{\mathrm{env}}$	$95\text{\,}\mathrm{K}$
$M$	$2.0\text{\,}\mathrm{kg}$	$T_{\mathrm{death}}$	$233.15\text{\,}\mathrm{K}$
$t_{\mathrm{night}}$	$354\text{\,}\mathrm{h}$	$C_{\mathrm{th}}$	$2000\text{\,}\mathrm{J}\text{\,}{\mathrm{K}}^{-1}$
$\Delta t$	$54.1\text{\,}\mathrm{s}$	$G_{\mathrm{sky}}$	$0.0108\text{\,}\mathrm{W}\text{\,}{\mathrm{K}}^{-1}$
$v_{\mathrm{cmd}}$	$0.015\text{\,}\mathrm{m}\text{\,}{\mathrm{s}}^{-1}$	$Q_{\max}$	$2.0\text{\,}\mathrm{W}$

TABLE I: Model parameters.

Fig. 2: Overview of the learning system. Left: fourteen local observation channels enter a 64-unit LSTM whose state persists across thermal steps; a skip connection passes the input directly to the output layer, which produces a velocity command. ES perturbs the shared parameters

\theta

and updates them from a single scalar returned after the full night. Middle: the simulated swarm, colored by sky exposure

\chi

. Right: the control loop deployed on every robot, using the trained

\theta

unchanged.

Training budget	Training budget	Resulting threshold	Static
( $\times$ bound)	(pcm-h)	(pcm-h)	survivors
1.10	151.9	155.8	14/20
1.18	163.0	155.7	14/20
1.40	193.3	176.4	19/20

TABLE II: Sensitivity of the survival threshold to the training budget. The budget is an input, the threshold an outcome; both are in pcm-h. One policy per budget, same seed.

Fig. 3: Emergence from a scattered start; all rows share one initial configuration. Color is each robot’s time-averaged exposure $\langle\chi\rangle$ up to that frame; crosses are dead robots. Under each frame: robots alive and the swarm’s mean $\langle\chi\rangle$ . The thresholds of Table III are a separate quantity, bisected on the budget rather than read off these runs. Top: the learned policy gathers within five hours and circulates, every average converges, 20/20 survive. Middle: the gather-only rule, which aggregates but never circulates, forms the same huddle, but a core–rim gradient persists all night and nine robots are lost. Bottom: under zero command the swarm stays scattered and all are lost by 225 h.

Condition	$n$	Threshold	$\times$ bound
Dispersed, no command	—	328.5	2.38
Motionless huddle	—	246.0	1.78
Gather-only rule	—	250.3	1.81
Learned, packed, shared ref.	3	$155.7\pm 1.3$	1.13
Learned, packed, no shared ref.	2	$154.3\pm 1.6$	1.12
Learned, scattered, shared ref.	2	$155.4\pm 0.6$	1.13

TABLE III: Survival thresholds in pcm-h. Learned entries pool converged runs.

Fig. 4: Transfer to larger swarms. Change in survival requirement relative to

N=20

, the size at which all five policies were trained; they were evaluated at

N=30

and

N=40

without retraining. Squares are the mean across policies with bars spanning best and worst; faint points are the individual policies. The grey region is the change available from the equalization bound.

Channel zeroed	s0	s1	s2
None	157.1	155.8	155.4
Neighbor stored energy	161.4	157.5	155.8
Neighbor temperature	179.8	154.9	159.6
Own sky exposure	157.1	159.2	242.6
Both neighbor channels	175.5	154.9	268.4

TABLE IV: Observation ablation results, reported per policy.

Swarm Robotics

Swarm robotics studies how groups of robots coordinate through local sensing, communication, and interaction to accomplish shared tasks.

See all 99 papers

Learning Thermal-Aware Locomotion Policies for an Electrically-Actuated Quadruped Robot
Mar 2, 2026Letian Qian, Yuhang Wan, Shuhan Wang +1Robot Locomotion Quadruped Locomotion
State Key Laboratory of Intelligent Manufacturing Equipment and Technology, Huazhong University of Science and Technology, Wuhan, China
Learning to Balance Motor Thermal Safety and Quadrupedal Locomotion Performance with Residual Policy
May 26, 2026Yuhang Wan, Weixian Lin, Letian Qian +5RL for Legged Locomotion Residual Policy Learning
School of Mechanical Science and Engineering, Huazhong University of Science and Technology, Wuhan, 430074, China
Self-organizing reinforcement learning: a learner-free swarm approaches the explore-exploit optimum
Oct 18, 2026Galen J. WilkersonMulti-Agent Resource Allocation Exploration-Exploitation Tradeoff
Komplexity AI LLC

Mar 2, 2026J/K move · Enter open · S save

Learning Thermal-Aware Locomotion Policies for an Electrically-Actuated Quadruped Robot

Letian Qian, Yuhang Wan, Shuhan Wang, Xin Luo

State Key Laboratory of Intelligent Manufacturing Equipment and Technology, Huazhong University of Science and Technology, Wuhan, China

Robot Locomotion Quadruped Locomotion RL for Legged Locomotion Constrained RL

Open paper AskPDF arXiv

Electrically-actuated quadrupedal robots possess high mobility on complex terrains, but their motors tend to accumulate heat under high-torque cyclic loads, potentially triggering overheat protection and limiting long-duration tasks. This work proposes a thermal-aware control method that incorporates motor temperatures into reinforcement learning locomotion policies and introduces thermal-constraint rewards to prevent temperature exceedance. Real-world experiments on the Unitree A1 demonstrate that, under a fixed 3 kg payload, the baseline policy triggers overheat protection and stops within approximately 7 minutes, whereas the proposed method can operate continuously for over 27 minutes without thermal interruptions while maintaining comparable command-tracking performance, thereby enhancing sustainable operational capability.

Abstract

Swarm Robotics

Explore similar work

Learning Thermal-Aware Locomotion Policies for an Electrically-Actuated Quadruped Robot

Learning to Balance Motor Thermal Safety and Quadrupedal Locomotion Performance with Residual Policy

Self-organizing reinforcement learning: a learner-free swarm approaches the explore-exploit optimum