DottyBot: Omnidirectional Bio-Assistive Kinematics & Architectural Blueprint
Modern clinical, therapeutic, and hospital facilities present acute spatial navigation challenges. Traditional differential-drive mobility systems and standard powered wheelchairs possess fixed wheel alignments, necessitating non-zero turning radii. In confined environments—such as crowded intensive care bays, tight residential washrooms, and high-density pharmaceutical cleanrooms—standard wheelchairs demand complex, multi-point turns that risk collisions with sterile equipment, patient beds, and clinical monitors. DottyBot simulates an advanced bio-assistive robotic platform engineered with an omnidirectional Mecanum wheel drive, coupling complete holonomic freedom with real-time sensory telemetry and sub-second Internet of Things (IoT) networking.
1. Kinematic Formulation of the 4-Wheel Mecanum Chassis
The core biomechanical and mechanical utility of DottyBot derives from its four independent Mecanum wheels. Each wheel features an array of passive, barrel-shaped peripheral rollers positioned at a fixed angle $\alpha = 45^\circ$ ($\pm \pi/4\,\text{rad}$) relative to the wheel’s circumferential rolling plane. When motor torque rotates the primary wheel hub, the passive rollers exert lateral traction forces against the ground surface. By independently modulating the rotational angular velocities of the four drive motors ($\omega_{FL}, \omega_{FR}, \omega_{BL}, \omega_{BR}$), longitudinal and lateral reaction force components can be constructively summed or destructively canceled, permitting the chassis to move in any planar direction without altering its physical orientation.
Let the platform's posture in global Cartesian space be represented by the vector $\mathbf{q} = [X, Y, \theta]^T$, where $(X, Y)$ is the planar geometric center of the chassis and $\theta$ denotes the heading angle. In the local robot coordinate frame, the instantaneous operational velocity vector is defined as:
$$\mathbf{v}_{chassis} = \begin{bmatrix} v_x \\ v_y \\ \omega_z \end{bmatrix}$$
where $v_x$ represents the lateral (strafing) velocity, $v_y$ represents the longitudinal (forward/reverse) velocity, and $\omega_z = \dot{\theta}$ represents the chassis rotational yaw rate. Let $r$ represent the nominal wheel radius, $2 l_x$ represent the longitudinal wheelbase (distance between the front and rear wheel axles), and $2 l_y$ represent the lateral track width (distance between the left and right wheel contact patches). The inverse kinematic model maps desired platform velocity to the four required wheel rotational angular speeds:
$$\begin{bmatrix} \omega_{FL} \\ \omega_{FR} \\ \omega_{BL} \\ \omega_{BR} \end{bmatrix} = \frac{1}{r} \begin{bmatrix} 1 & -1 & -(l_x + l_y) \\ 1 & 1 & (l_x + l_y) \\ 1 & 1 & -(l_x + l_y) \\ 1 & -1 & (l_x + l_y) \end{bmatrix} \begin{bmatrix} v_y \\ v_x \\ \omega_z \end{bmatrix}$$
Direct inspection of this inverse kinematic Jacobian matrix reveals the mechanical principles governing specialized maneuvers:
- Pure Longitudinal Translation ($v_y \neq 0, v_x = 0, \omega_z = 0$): All four wheels rotate synchronously in the same direction ($\omega_{FL} = \omega_{FR} = \omega_{BL} = \omega_{BR} = v_y / r$). The lateral reaction forces generated by the opposing $45^\circ$ roller angles cancel out symmetrically, producing pure forward thrust.
- Pure Lateral Strafing ($v_y = 0, v_x \neq 0, \omega_z = 0$): Diagonal pairs rotate in mutually reverse directions ($\omega_{FL} = -\omega_{FR} = -\omega_{BL} = \omega_{BR} = -v_x / r$). Forward thrust forces cancel entirely, while the roller contact vectors sum along the transverse axis, propelling the platform directly sideways without any yaw rotation.
- Zero-Radius Turning ($v_y = 0, v_x = 0, \omega_z \neq 0$): The wheels on the right flank rotate in direct opposition to those on the left flank ($\omega_{FL} = \omega_{BL} = -(l_x + l_y)\omega_z / r$, while $\omega_{FR} = \omega_{BR} = (l_x + l_y)\omega_z / r$), creating a pure couple moment about the center of mass.
2. Sensory Telemetry & Autonomous Clinical Safety Envelopes
In bio-assistive and clinical mobility applications, passive omnidirectional capability must be paired with proactive spatial awareness. DottyBot incorporates a multi-modal sensor telemetry pipeline designed to safeguard impaired users, vulnerable patients, and sensitive medical instruments:
- Time-of-Flight Ultrasonic Collision Radar: Mounted along the leading edge of the chassis, the ultrasonic transducer emits high-frequency acoustic waves ($40\,\text{kHz}$) through an active conical beam ($30^\circ$ solid angle). In this simulator, continuous geometric raycasting resolves the nearest boundary distance $d_{obst}$. When approaching clinical infrastructure (such as hospital beds, IV drip stands, or walls), an automated deceleration envelope triggers: if $d_{obst} \le d_{crit}$ (configured at $25\,\text{cm}$), the control loop engages regenerative braking override to eliminate impact risk.
- Acoustic Ambient Detection: Clinical corridors and intensive care units require strict auditory noise mitigation to preserve patient circadian rest. DottyBot monitors ambient noise thresholds via a digital sound transducer, logging sound pressures to alert personnel if an onboard acoustic alarm is sounded or if ambient noise exceeds clinical thresholds.
- Thermal Diagnostic Telemetry: Assisting in hospital logistics often involves transporting biological reagents, blood samples, or pharmaceutical infusions. Live temperature logging verifies transport container integrity along delivery routes.
3. IoT Network Topology, ESP32-S3 Micro-Server & AbortController Override
DottyBot's physical hardware prototype is driven by an Arduino Uno R4 WiFi, incorporating an onboard ESP32-S3 dual-core microcontroller operating in local Access Point (AP) Mode. The ESP32 broadcasts a private, WPA2-encrypted Wi-Fi local network, serving this complete HTML5/JavaScript dashboard directly to mobile diagnostic handsets without requiring external internet infrastructure.
High-frequency HTTP telemetry loops run the risk of saturating the microcontroller's lightweight TCP/IP stack. During early physical testing, continuous analog joystick polling dispatched hundreds of simultaneous asynchronous HTTP requests, causing packet queueing that delayed safety-critical stop commands by up to 1.8 seconds. This latency was solved by implementing two client-side architectural patterns:
- Deadband & State-Change Filtering: Kinematic vector recalculations occur locally at 60 Hz, but network packets are dispatched only when discrete direction vectors switch or when velocity increments exceed a $10\%$ deadband threshold.
- AbortController Mid-Air Request Cancellation: The emergency brake routine instantiates a global JavaScript
AbortController. Triggering the Emergency Stop button immediately aborts all in-flight telemetry and sensor polling requests, ensuring the physical /stopall halt packet takes instant precedence across the hardware communication bus.
4. How to Use This Interactive Laboratory
The interactive workspace above provides real-time holonomic simulation and hardware diagnostics:
- Interactive Canvas Arena: Click and drag the DottyBot chassis directly within the hospital ward arena to relocate it, or drag the cyan Target Puck to set an autonomous waypoint. When Puck Navigation Mode is active, the robot continuously drives its holonomic Jacobian solver toward the target coordinate.
- Analog Joystick & Discrete D-Pad: Drag the central joystick handle to produce continuous $360^\circ$ omnidirectional force vectors, or switch to the Discrete D-Pad for cardinal and diagonal heading increments.
- Kinematic Speed Profile: Switch between LOW (Clinical Safe Mode, 0.35 m/s) for navigating delicate patient bedsides and HIGH (Transit Mode, 1.0 m/s) for long-distance hospital corridor transit.
- Autonomous Kinematic Sequencer: Execute pre-programmed choreographies (such as the Ward Perimeter Box or Bedside Zero-Radius Pivot) to observe exact Mecanum roller wheel speeds during automated clinical delivery routines.
Explore the BioniChaos Ecosystem
Open Access License: This interactive educational module is released under
CC BY-NC 4.0 (Attribution-NonCommercial)
for non-commercial research, academic study, and clinical education.
Commercial & Enterprise Licensing: For white-labeling, proprietary LMS/course embedding, hardware dashboard telemetry integration, or custom feature engineering, secure a commercial license at
BioniCloud.com or contact
Dr. Yuri Beno.