Hexapod robots can achieve static stability with fewer actuated joints than bipeds or quadrupeds, yet many platforms still use 3-DOF legs, increasing weight and continuous torque requirement with limited gain in locomotion capability on flat, inclined and moderately rough terrains. We release Spiderbot, an open-source hexapod that uses a 4-bar linkage with a passive spring to mechanically support body weight, with a 2-DOF per-leg design that substantially reduces energy consumption. This mechanism substantially offloads gravitational torque during standing stance consuming only 1.5W (reduction of over 90% over the unsprung version and up to 96% over other similar hexapods). The passive spring compensation extends to payloads of up to 3.25kg with no additional torque requirements. The platform enables long-duration deployments on a modest battery budget and costs under $400, making it suitable for large-scale multi-agent experiments. We validate the locomotion capabilities of the platform with an RL policy trained in mjlab, including successful sim-to-real transfer, despite the complexity of the mechanism. The platform is evaluated on flat and rough terrains, slope up to 15∘ and step obstacles. We release all the CAD files, assembling instructions, and full training and deployment code along with the model checkpoints at https://erc-bpgc.github.io/SpiderBot/.
Figures & tables
Fig. 1: Spiderbot, the proposed open-source 12-DOF hexapod platform featuring spring-assisted four-bar legs for passive gravity compensation.
Fig. 2: Foot workspace visualisation
Fig. 3: 2-DOF vs 3-DOF leg workspace comparison
Fig. 4: Leg design and dimensions for our hexapod (measurements in mm).
Fig. 5: Free Body Diagram for link 1 (femur).
Symbol
Value
Description
m1
0.0608kg
Mass of Link 1 (L1)
m2
0.0125kg
Mass of Link 2 (L2)
m3
0.0124kg
Mass of Link 3 (L3)
mhb
2.59kg
Mass of base Body (B)
g
9.81m/s2
Gravitational acceleration
lL1
0.1463m
Length of L1
TABLE I: Known physical quantities of the robot
Fig. 6: Total Robot Torque Requirement (Femur Joint) vs Spring Stiffness (k)
Fig. 7: Additional torque requirement as a function of spring stiffness and payload mass for neutral stance configuration
Component
Qty.
Total Cost (USD)
ST3215 High-Torque Serial Servo Motor
6
126
SC15 Compact Serial Servo Motor
6
108
Arduino Uno Q Development Board
1
44
PLA Filament Spool (1 kg)
1
10
2-Cell 5200 mAh Li-Po Battery
1
30
RC Shock Spring 450 N/m
12
18
TABLE II: Bill of Materials (BOM) for Spiderbot
Fig. 8: Electronics architecture of Spiderbot featuring compute, sensors and actuation.
Parameter
Range
Operation
Foot friction coefficient
[0.3,1.2]
Absolute
Encoder bias (rad)
[−0.015,0.015]
Absolute
Base CoM offset (m)
[−0.025,0.025]xy,[−0.03,0.03]z
Additive
Base link mass (kg)
[−1.0,1.0]
Additive
Limb mass scale
[0.8,1.2]
Scaled
Motor zero offset (rad)
[−0.035,0.035]
Additive
TABLE III: Domain randomization parameters applied at training startup.
Terrain Type
Proportion
Parameters
Flat
0.20
Baseline flat ground.
Pyramid Stairs
0.20
Step height: 2–5 cm, step width: 0.5 m.
Inverted Stairs
0.20
Step height: 2–5 cm, step width: 0.5 m.
Pyramid Slope
0.10
Slope gradient: 0.05–0.35.
Inverted Slope
0.10
Slope gradient: 0.05–0.35, inverted.
Random Rough
0.10
Uniform noise height: 1.5–5 cm.
TABLE IV: Terrain types used during training.
Category
Reward Term
Weight
Task
Track linear velocity
+5.00
Track angular velocity
+2.00
Regulation
Upright orientation
+1.00
Joint pose
+1.00
Penalty
Joint position limits
−1.00
Action rate (L2)
−0.10
TABLE V: Active reward terms and their weights.
Fig. 9: Different environments in MuJoCo simulation. From left to right: flat ground, rough terrain, stairs
Energy-efficient robot locomotion studies how robot mechanics and control can reduce power use and cost of transport across terrains and movement modes.