Design of an In-Pipe Robot with Contact-Angle-Guided Kinematic Decoupling for Crosstalk-Suppressed Locomotion
Authors: Min Yang, Yang Tian, Longchuang Li, Jun Ma, Shugen Ma
Organizations: Robotics and Autonomous Systems Thrust, The Hong Kong University of Science and Technology (Guangzhou), Guangzhou 511453, China · Department of Engineering, Shinshu University, Nagano 380-8553, Japan · College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China
In-pipe inspection robots must traverse confined pipeline networks with elbows and three-dimensional fittings, requiring both reliable axial traction and rapid rolling reorientation for posture correction. In compact V-shaped platforms, these functions often rely on shared contacts or indirect actuation, which introduces strong kinematic coupling and makes performance sensitive to geometry and friction variations. This paper presents a V-shaped in-pipe robot with a joint-axis-and-wheel-separation layout that provides two physically independent actuation channels, with all-wheel-drive propulsion and motorized rolling reorientation while using only two motors. To make the decoupling mechanism explicit and designable, we formulate an actuation transmission matrix and identify the spherical-wheel contact angle as the key geometric variable governing the dominant roll-to-propulsion leakage and roll-channel efficiency. A geometric transmission analysis maps mounting parameters to the contact angle, leakage, and efficiency, yielding a structural guideline for suppressing crosstalk by driving the contact angle toward zero. A static stability model further provides a stability-domain map for selecting torsion-spring stiffness under friction uncertainty to ensure vertical-pipe stability with a margin. Experiments validate the decoupling effect, where during high-dynamic rolling in a vertical pipe, the propulsion torque remains nearly invariant. On a multi-material testbed including out-of-plane double elbows, the robot achieved a 100% success rate in more than 10 independent round-trip trials.
Figures & tables
Fig. 1: Prototype of the proposed V-shaped in-pipe robot enabled by a joint-axis-and-wheel-separation architecture.
Fig. 2: Conceptual comparison of architectures: (a) Coupled motion vs. (b) Our decoupled motion.
Fig. 3: 3D CAD model of the proposed in-pipe robot. The system integrates an AWD propulsion train and a centrally actuated spherical wheel for rolling reorientation.
Method
P channel
R channel
P & R co-available?
How the P–R conflict is resolved (core mechanism)
Ours
Dedicated AWD propulsion motor
Dedicated central roll motor
Yes (with two actuators)
Physical separation (joint-axis-and-wheel-separation) enables direct center roll torque; residual roll → propulsion leakage is made explicit and suppressible via geometry (contact angle)
Mode switching by middle-joint rotation direction; rolling is largely contact-mediated (environment reaction) rather than an independent commanded roll-torque channel
[ 24 ] (2021)
Underactuated + one-way clutch constraint
Mode-select via clutch/direction
No (mode-switch)
Clutch-constrained separation : switching between motion modes by selecting motor drive direction (not simultaneous independent P/R)
TABLE I: Key architectural differences to representative close prior art V-shaped/articulated in-pipe robots at the actuation-allocation level.
Fig. 4: Geometric constraints and end-view configuration of the V-shaped robot inside a circular pipe. (a) Pipe cross-section schematic defining the pipe diameter Dp , the omni-wheel pair spacing Wo , the wheel radii Ro (omni wheel) and Rs (spherical wheel), the pipe center O , and the nominal omni-wheel-to-wall geometric clearance Hwp0 shown in the equivalent 2-D model. (b) End-view configuration used for the geometric constraint and contact-angle analysis: the core structural design parameters (L1,L2,a,b,n) determine the derived configuration angles (α1,α2,θ1,θ2) and the total V-opening angle θ=θ1+θ2 . The distance Hs denotes the offset between the omni-wheel centerline and the spherical-wheel center along the pipe diameter direction (as indicated). At the spherical-wheel–pipe interface, α0 marks the angular location of the nominal (ideal) contact point on the pipe wall, while α denotes the actual contact angle used later to characterize transmission crosstalk.
Fig. 5: Unified end-view model and notation used in both the static stability analysis (Section III) and the actuation transmission analysis (Section IV). The model defines the wheel–pipe normal forces FN0,FN1,FN2 at the spherical wheel (index 0) and the two omni-wheel pairs (indices 1 and 2), the corresponding friction forces f0,f1,f2 , the gravitational loads Gk of the main modules (indexed in the figure), and the joint torque MJ at the V-arm pivot. The inset illustrates the spherical-wheel contact and defines the contact angle α as the deviation between the direction of the actual contact force/velocity transmission and the ideal circumferential direction for pure rolling about the pipe axis; the rolling torque τr and the induced axial-leakage component τp are also illustrated.
Specification
Value
Length of Link 1 ( L1 )
105 mm
Length of Link 2 ( L2 )
75 mm
Total Length
240 mm
Total Mass
750 g
Omni-wheel Radius ( Ro )
30 mm
Spherical-wheel Radius ( Rs )
22.5 mm
TABLE II: Key specifications of the robot module
Fig. 6: Effect of link length ratio on normal force distribution. In this analysis, the length of link L1 is fixed at 105 mm, and different L2/L1 ratios are obtained by varying the length of link L2 . All other design parameters are taken from the nominal values in Table II . The gray dashed lines represent the length of link actual values in Table II .
Fig. 7: Stability-domain analysis showing the effect of torsion-spring stiffness K and spherical-wheel friction coefficient μs on anti-gravity feasibility. (a) Three-dimensional stability surface, where the stability safety margin is defined as S=fmax,total/Gtotal . When the surface is above the red reference plane S=1 , the system is stable against gravity in a vertical pipe. (b) Two-dimensional stability-domain plot, obtained by projecting (a) onto the (μs,K) plane. The white region indicates stable operation ( S≥1 ) and the red region indicates slipping ( S<1 ). The thick red curve is the critical boundary at S=1 , with K reported in N⋅mm/deg .
Fig. 8: Contact-angle surface α(a,n−b) computed from the geometric constraints (Eq. ( 5 )). The marked point corresponds to the prototype mounting parameters.
Fig. 9: Inverse-dynamics simulation of coupling effects on required motor commands. (a) Prescribed high-dynamic roll trajectory with axial station-keeping against gravity. (b) Required propulsion command τp and roll command τr for three transmission configurations: geometry-derived non-optimized baseline (zero separation), optimized design, and a coupling stress test with imposed normalized leakage ratios.
Fig. 10: Platform for experiments in this work.
Fig. 11: System architecture and data flow. Operator commands are transmitted wirelessly via a 2.4G link from the handle to the onboard receiver and processed by the microcontroller. Motor control is executed via a CAN Bus. System telemetry is sent back to a laptop for real-time monitoring via a separate wireless data link. The main power is supplied through a cable and an electric slip ring to allow for unrestricted robot rotation.
Fig. 12: Experimental validation of kinematic decoupling under gravitational load. The robot was held stationary in a vertical pipe while the rolling motor was commanded to perform five continuous revolutions (from 5.56 s to 19.76 s ). Despite the high and dynamic torque from the rolling motor (orange), the propulsion motor’s torque (green) remained exceptionally stable with a standard deviation of less than 0.0077 N⋅m , demonstrating minimal cross-talk between the two axes.
Fig. 13: Kinematic sequence demonstrating the “functional slip” mechanism during a vertical pipe test. While the robot performs continuous rolling (full sequence not shown), it exhibits a slow, controlled axial slip due to gravity. The slip distance was approximately 20 mm (slip distance 1) after the first revolution (5.6 s ) and accumulated to 95 mm (slip distance 2) after four revolutions (17.0 s ). This result validates our “robustness-by-design” strategy, where a potential static failure (stalling) is transformed into a predictable, non-critical motion.
Fig. 14: Kinematic sequence of the robot traversing the OPDE. (a) Ascending pass. (b) Descending pass. The continuous motion and change in roll angle demonstrate the concurrent execution of propulsion and rolling.
Fig. 15: Demonstration of posture-adaptive correction in the vertical elbow (WE4). (a) Ascending pass. (b) Descending pass. The robot enters with a visible misalignment but rapidly self-corrects its posture within 0.2 s, showcasing the dynamic response of the decoupled rolling system.
Pass Direction
Traversal Section
Gravitational Effect
Length ( m )
Time ( s )
Avg. Speed ( m/s )
Forward Pass
1. CB ( Acrylic )
Assist ( +g )
2.60
12.0
0.22
2. WE 1, 2, 3 (Eng. Plastic)
Horizontal ( ∼0g )
0.47 (avg)
2.1 (avg)
0.22
3. OPDE (Eng. Plastic)
Inhibit ( −g )
0.94
8.1
0.12
4. WE 4 (Eng. Plastic)
Inhibit ( −g )
0.47
1.6
0.29
5. WE 5 (Eng. Plastic)
Horizontal ( ∼ 0 g )
0.47
1.8
0.26
Backward Pass
6. WE 5 (Eng. Plastic)
Horizontal ( ∼ 0 g )
0.47
1.7
0.28
TABLE III: Performance Analysis of Bidirectional Traversal Under Varying Gravitational Effects (Segment-wise time/speed from one randomly selected trial among 10 teleoperated round-trip trials)
This paper proposes, an improved version of the 3 in sewer pipe inspection robot equipped with an emergency evacuation mechanism. The low traction force and poor stepover capability, which were challenges of the first version, have been improved by simply connecting the propulsion units. The coupled propulsion units feature a differential mechanism capable of posture changes via a single wire, enabling adaptation to pipe diameter variations. To traverse obstacles like pipe joints, a control method was devised that detects obstacle contact through current load on the drive wheel motors and slackens the wire. This method was verified through simulated pipe experiments. Load comparisons were made using current waveforms applied to the drive wheels. Our proposed control method significantly improved the step-over capability of the new articulated robots.
Shoya Umemura, Ryota Taniguchi, Atsushi Kakogawa
Department of Robotics, Faculty of Science and Engineering, Ritsumeikan University, 1-1-1 Noji-higashi, Kusatsu, Shiga, 525-8577, JAPAN
An asymmetric two-link robot supported atop a flat platform by wheels that roll and pivot freely, but do not slip laterally, will develop forward momentum if the joint between the links is actuated internally. In particular, oscillations in the joint angle will generate undulatory locomotion suggesting fishlike swimming. If two such robots surmount a common platform that's free to translate with its own inertial dynamics, then the individual robots' dynamics will be coupled so that the locomotion of either robot is affected by that of the other. We develop a mathematical model for this system and present simulations demonstrating its behavior. We then consider a single robot with an unactuated joint rolling atop a platform that moves under control, and show that actuation of the platform is sufficient to dictate the robot's behavior. In particular, with the acceleration of the platform as an input, the robot's heading can be made to track a chosen function of time. This is sufficient to guarantee that the robot can be induced to orbit a fixed point on the platform or to locomote persistently in a desired direction.
Hamidreza Moradi, Scott David Kelly
Department of Mechanical Engineering and Engineering Science, University of North Carolina at Charlotte, Charlotte, NC 28223 USA
Exploration of confined spaces, such as pipelines and ducts, remains challenging for conventional rigid robots due to limited space, irregular geometry, and restricted access. Inspired by caterpillar locomotion and sensing, this paper presents a compact spring-based tendon-driven continuum robot that integrates with commercial robotic arms for confined-space inspection. The system combines a mechanically compliant continuum body with a tendon actuation module, enabling coupled bending and axial length change, and uses a constant-curvature kinematic model for positional control. Experiments show a mean position error of 4.32 mm under the proposed model and control pipeline. To extend the system from motion to inspection, we integrate an artificial bristle contact sensor and demonstrate surface perception and confined-space exploration through contact interactions. This compact and compliant design offers a cost-effective upgrade for commercial robots and promises effective exploration in challenging environments.
Zhixian Hu, Yu She, Juan Wachs
Edwardson School of Industrial Engineering, Purdue University, West Lafayette, IN, USA.