cs.ROSep 28, 2026

Simulation for Planetary Robotic Perception and Autonomy: A Concise Survey of Recent Capabilities and Gaps

Authors: Hoyun Kim, Giseop Kim

Organizations: Department of Robotics and Mechatronics Engineering, DGIST, Daegu, Republic of Korea

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Planetary robotics is an important enabler of scientific exploration in environments where direct human-in-the-loop operation is costly, hazardous, or infeasible. However, developing and validating planetary robotic systems remains difficult because representative field testing is expensive, limited, and often unrepeatable under mission-relevant conditions. In this setting, simulation serves as a central tool for perception and autonomy research, synthetic data generation, system integration, and pre-deployment evaluation. Despite its importance, the literature on planetary robotics simulation remains dispersed across different simulation engines, implementations, and application settings. This paper surveys simulation works for planetary robotic perception and autonomy across four practical axes: Openness and Availability, Scenario and Platform Coverage, Sensor and Perception Support, and Environmental and Operational Realism. The surveyed simulation works report visual or physical fidelity and support perception-oriented workflows. They also indicate uneven public availability, rover-centered coverage, partial support for specialized sensing modalities, and uneven reporting of operational constraints such as onboard computation, energy, and communication restrictions.

Figures & tables

Fig. 1 : Structure of this survey. The paper first derives planetary simulation requirements, then surveys twenty-four simulation works along four analysis axes, summarizes the resulting gaps as open challenges, and provides practical starting points for new users.
Fig. 2 : Chronological roadmap of the surveyed planetary simulation works as domain lanes (top: lunar; middle: cross-domain or other bodies; bottom: Mars). Each lane starts from an early mission-era anchor (Apollo LM mission simulators, the ROAMS rover simulator at NASA JPL, and selected development and release milestones of general-purpose engines; shown as uncited historical markers), and the wavy break compresses the intervening decades before the recent time series. Marker shape encodes the platform or task category, and marker fill encodes simulation resource availability (filled: public release; outline: no release identified; half: on request). Further details on the surveyed works are provided in Table I . Dark-red squares mark major space-mission events and open gray squares mark simulation-engine milestones.
Fig. 3 : Key factors distinguishing planetary from terrestrial autonomy and their implications for simulation design. Each domain factor (left) is linked to its reported planetary characteristic (center) and the simulation capability it implies (right), grouped into physical environment, operational constraints, and robotic system scope.
WorkYearDomainCategoryEngine / ToolROSAccessTerrain SourceSensors / Outputs
Giubilato et al. [ 25 ]2020MarsRoverGazeboROSP / CRealSt Cam, 3D LiDAR
Toupet et al. [ 29 ]2020MarsRoverROS-basedROSP / –SimulatedSt Cam, rover state
OAISYS [ 8 ]2021Planet.Multi-platformBlender–P / CSyntheticRGB-D (extensible)
GraspPlanetary [ 10 ]2022LunarManipulationGazebo + DARTR2P / CProceduralDepth/intensity, octrees
MarsSim [ 26 ]2023MarsRoverGazeboROSP / –RealMast-mounted Cam
SEELO [ 11 ]2023LunarExcavatorUnity–P / –RealVision, IMU
TABLE I : Overview of the twenty-four surveyed works. Access is paper / simulation resources : P = public publication page or full-text link; C = work-specific code; C* = partial components or evaluation code; B = simulator executable; D = data/assets; R = request channel; – = no release identified. R1/R2 = ROS 1/2; ROS = supported, version unspecified. St = Stereo; Sem = Semantic; Seg = Segmentation; Enc = Wheel encoder; Alt = Altimeter; Inc = Inclinometer.
Fig. 4 : Distribution of surveyed simulation works by (a) access to work-specific simulation resources and (b) primary platform or task relevance. Public releases include code, partial components, executables, and data (Table I ).
Sensing ModalitySurveyed WorksMission / Prototype
Mono / Stereo Camera✓✓ [ 46 ]
RGB-D✓✓ [ 47 ]
2D / 3D LiDAR✓✓ [ 42 ]
IMU✓✓ [ 46 ]
Wheel Encoder / Odometry✓ [ 31 , 12 , 11 ]✓ [ 42 ]
Semantic / Instance Labels✓–
TABLE II : Sensing modality coverage reported in the surveyed works and in mission, planned-mission, prototype-stage, or agency research-stage literature. Entries indicate reported support or use rather than comprehensive coverage of the field or baseline simulator requirements.
TABLE III : Reported treatment of communication, onboard computational resource limits, and energy in the surveyed works.
Fig. 5 : Mission-scene view of how environmental and operational realism factors appear in planetary simulation. The upper image provides the mission context 2 2 2 The Mars scene image in Fig. 2 was generated using ChatGPT Image 2.0. , while the indexed blocks give an at-a-glance summary of the implication and reporting gap associated with each factor. Table III provides the source-based operational comparison and distinguishes explicit models from limited treatments.

Explore similar work

CardsList

  1. MarsLab: A Martian Rover Simulator for Planetary Rover Autonomous Navigation

    Sep 28, 2026Hoyun Kim, Beomsu Kim, Giseop KimAutonomous Surface VehiclesTerrain

    Department of Robotics and Mechatronics Engineering, DGIST, Daegu, Republic of Korea

  2. PerceptTwin: Semantic Scene Reconstruction for Iterative LLM Planning and Verification

    Jun 2, 2026Charlie Gauthier, Sacha Morin, Liam PaullLarge Language Model PlanningRobotic Perception

    Department of Computer Science and Operations Research, Universit´e de Montr´eal, Montr´eal, QC, Canada. · Mila - Quebec AI Institute, Montr´eal, QC, Canada. · CIFAR AI Chair.

  3. OmniPlanner: Universal Exploration and Inspection Path Planning Across Robot Morphologies

    Mar 4, 2026Angelos Zacharia, Mihir Dharmadhikari, Mohit Singh +1Robot PlanningAerial Robots

    Department of Engineering Cybernetics, Norwegian University of Science and Technology (NTNU), 7034 Trondheim, Norway